Below every paper are TOP 100 most-occuring words in that paper and their color is based on LDA topic model with k = 7.
(It looks like 0 = graphical learning?, 1 = reinforcement learning, 2 = deep learning, 3 = non-parametrics?, 4 = matrix factorization?, 5 = neuroscience, 6 = optimization)
Toggle LDA topics to sort by:
TOPIC0
TOPIC1
TOPIC2
TOPIC3
TOPIC4
TOPIC5
TOPIC6
Also note that you can filter by the day each poster appears: click on the day next to a paper to filter by that day.
Time--Data Tradeoffs by Aggressive Smoothing
John J. Bruer, Joel A. Tropp, Volkan Cevher, Stephen Becker
John J. Bruer, Joel A. Tropp, Volkan Cevher, Stephen Becker
[number, maximal, random, apply, increase, fact, experiment, set, procedure] [algorithm, proposition, total, case, result, bound, optimal, order, phase, sequence] [work, based, learning, applied, geometry, achieve, accurate, additional] [sample, figure, parameter, average, measurement, supplemental, data, amount, gaussian] [recovery, matrix, sparse, vector, dimension, exact, condition, sparsity, norm, rank, suggests, siam, numerical, ambient, point, nonzero] [inverse, signal, panel, observed] [problem, smoothing, statistical, size, convex, cost, method, dual, rlip, normalized, linear, regularizer, solve, primal, computational, error, excess, smoothed, aggressive, constant, optimization, descent, regularized, assume, gradient, geometric, convergence, large, complexity, relative, min, arg]
Bounded Regret for Finite-Armed Structured Bandits
Tor Lattimore, Remi Munos
Tor Lattimore, Remi Munos
[number, asymptotically, structure, interval] [regret, arm, bound, algorithm, optimal, ucb, bandit, expected, return, ern, logarithmic, action, lower, case, proof, thompson, best, setting, lai, general, lemma, max, chosen, gap, cumulative, choose, optimistic, bubeck, suffers, depend, leung, learner, minimum, unknown, agrawal, tze, bastien, choosing, bounded, herbert, upper, met, studied, advertising, main] [structured, work, learning, bounding, unit] [log, figure, parameter, true, sampling, sampled, expectation, robbins] [improved, second, union, analysis, small] [depicted] [theorem, problem, assume, function, constant, large, example, machine, note, suppose, stochastic, complexity]
Best-Arm Identification in Linear Bandits
Marta Soare, Alessandro Lazaric, Remi Munos
Marta Soare, Alessandro Lazaric, Remi Munos
[set, number, needed, identify, selection, consider, discrete, high] [arm, allocation, strategy, stopping, design, optimal, static, best, phase, bandit, oracle, algorithm, logn, conference, budget, bound, max, notice, case, reward, mab, adaptive, unknown, previous, pulling, suboptimal, hlb, study, space, discarding, criterion, expected, sequence, setting, gap, focus, difference, learner, studied, minimum] [learning, performance, well, additional] [sample, estimate, parameter, uncertainty, sampling, international, allows, figure] [condition, corresponding, cone, matrix, point] [directly, experimental, observed] [complexity, linear, problem, min, arg, empirical, direction, theory, implemented, objective, estimation, discard, term, quantity, corresponds, machine]
Learning to Optimize via Information-Directed Sampling
Dan Russo, Benjamin Van Roy
Dan Russo, Benjamin Van Roy
[set, random, ratio, entropy, consider, binary, paper, experiment, solution, high] [action, thompson, regret, optimal, ucb, algorithm, expected, bandit, reward, bound, probability, gain, upper, type, period, studied, customer, policy, chosen, online, assortment, cumulative, exploration, general, proposition, chooses, outcome, lower, van, russo, focus, dramatically, version, satisfy, search, minimizes, utility] [performance, work, learning, feature, applied, learn] [sampling, distribution, posterior, gaussian, bayesian, true, prior, figure, parameter] [product, knowledge, numerical, analysis, vector, establish, denote, greedy] [time, independent, mutual, drawn, presented] [linear, problem, optimization, bayes, simple, measure, theoretical, stochastic, strong, gradient, arg, computational, positive]
Efficient learning by implicit exploration in bandit problems with side observations
Tomáš Kocák, Gergely Neu, Michal Valko, Remi Munos
Tomáš Kocák, Gergely Neu, Michal Valko, Remi Munos
[graph, set, number, directed, consider, random, partial] [algorithm, learner, exploration, regret, action, observability, implicit, combinatorial, lemma, news, observe, setting, feedback, online, bandit, bound, side, alon, decision, auer, fpl, chosen, upper, access, mannor, shamir, neu, proof, adaptive, result, optimistic, total, guarantee, choose, main, clickthroughs, observes, vtt, previous] [learning, performance, based, propose, computer, proposed] [log, observation, scheme, distribution, figure, model, full, parameter, independence, estimate, prior, computationally, sampling, bias] [vector, requires, analysis, refer, running] [system, time, presented] [loss, problem, assume, theorem, optimization, framework, prove, variant]
Estimation with Norm Regularization
Arindam Banerjee, Sheng Chen, Farideh Fazayeli, Vidyashankar Sivakumar
Arindam Banerjee, Sheng Chen, Farideh Fazayeli, Vidyashankar Sivakumar
[consider, set, random, ball, characterization, covering, existing] [design, bound, result, probability, case, correlated, setting, literature, depend, upper, considered, general, focus, satisfy, proof, play] [unit, work, structured, based] [gaussian, restricted, parameter, variety] [norm, analysis, isotropic, matrix, suitable, condition, rsc, recovery, establish, anisotropic, cone, row, eigenvalue, special, subgaussian, vector, elementwise, characterize, equivalent, role, technique, denotes, xij, implying] [independent, width, noise, dependent, three] [error, estimation, inf, theorem, loss, linear, regularized, inequality, generalized, constant, constrained, note, precise, considers, convex, measure, size, geometric, function, form, assume, annals, regularization, including, term]
Bayesian Inference for Structured Spike and Slab Priors
Michael R. Andersen, Ole Winther, Lars K. Hansen
Michael R. Andersen, Ole Winther, Lars K. Hansen
[structure, experiment, consider, number, solution, marginal, apply, fact, provide] [algorithm, order, conference, interest, expected, oracle] [structured, proposed, joint, pattern, reconstruction, ieee, multiple, based, level, work] [prior, exponential, true, covariance, slab, model, bayesian, posterior, figure, gaussian, ber, approximate, inference, expectation, distribution, measurement, nmse, scheme, kernel, propagation, undersamplingsratio, cavity, allows, standard, approach, sampled, priori, aeeg, process] [sparsity, approximation, support, diagonal, sparse, matrix, formulation, rank, numerical, knowledge, synthetic, vector] [signal, spike, correlation, inverse, lars, snr, independent, recall] [function, update, group, squared, linear, problem, active, method, form, journal, framework, term, computational, example]
Deterministic Symmetric Positive Semidefinite Matrix Completion
William E. Bishop, Byron M. Yu
William E. Bishop, Byron M. Yu
[set, random, exists, minimal, collection, provide, unique, overlapping, corollary, solution, ordering, number] [algorithm, deterministic, bound, case, proof, previous] [overlap, work, reconstruction, learning, learn, original] [sampling, estimate, mask, full, covariance, step, scheme, figure, generate] [matrix, principal, rank, submatrices, spsd, completion, recovery, submatrix, noisy, noiseless, diagonal, indexed, exact, recover, row, exactly, success, pairwise, randomly, corresponding, low, ordered, denote, corrupted, revealed, orthonormal, establishes, reconstruct, symmetric, formed, necessity] [neural, observed, noise, estimated, indicate, three] [theorem, error, problem, block, assume, machine, respect, method, min, hold, positive, linear, assumption, dependence, generalized]
Efficient Sampling for Learning Sparse Additive Models in High Dimensions
Hemant Tyagi, Bernd Gärtner, Andreas Krause
Hemant Tyagi, Bernd Gärtner, Andreas Krause
[set, arbitrary, solution, high, consider, interval, program, stated, feasible, provide, subset, exist, degree] [algorithm, uniform, unknown, setting, bounded, choose, order, search, bound] [learning, propose] [log, estimate, sampling, step, sample, parameter, scheme, figure, gaussian, measurement] [noisy, point, additive, spline, approximation, recovery, estimating, sparse, small, robust, corrupted, recover, norm, close, univariate, analyze, denote, condition, component, involves, compressive, vector, recovers] [noise, assumed, external, individual, choice, account] [error, cubic, theorem, problem, linear, function, query, active, derive, note, convex, size, assumption, returned, large, smoothing, quadratic, smooth, theory, estimation, form, assume, stochastic, annals, theoretical]
A Drifting-Games Analysis for Online Learning and Applications to Boosting
Haipeng Luo, Robert E. Schapire
Haipeng Luo, Robert E. Schapire
[potential, set, consider, number, adversary, existing, high, continuous] [algorithm, minimax, online, hedge, regret, drifting, optimal, exp, general, player, robert, game, setting, appendix, lemma, adaptive, round, probability, normalhedge, yoav, bandit, strategy, difference, designing, chip, best, conference, bound, action, simply, case, appropriate, upper, study] [learning, training, proposed, computer] [distribution, processing, standard, exponential, step] [analysis, recover, exact] [property, fraction, time, derived, neural, simultaneously] [loss, convex, theorem, boosting, function, machine, generalized, variant, derive, dependence, problem, rate, simple, weak, optimization, error, margin, journal, constant]
Active Regression by Stratification
Sivan Sabato, Remi Munos
Sivan Sabato, Remi Munos
[random, label, number, consider, partition, ratio, provide, marginal, set, potential] [learner, optimal, algorithm, oracle, probability, lemma, bound, general, setting, conference, minimax, design, appendix, goal, guarantee, explicit] [learning, labeled, propose, based, improvement, input] [regression, sample, distribution, data, approach, model, sampling, parametric, prior, heteroscedastic, standard] [denote, condition, support, second, improved, provided, matrix, knowledge] [drawn] [active, risk, rate, passive, convergence, linear, function, theorem, predictor, statistical, suppx, example, squared, note, error, journal, assume, complexity, kanamori, loss, dependence, asymptotic, machine, sugiyama, denoted, depends, hsu, estimator, rejection, hold, equality, inequality, size, assumption, generalized]
Greedy Subspace Clustering
Dohyung Park, Constantine Caramanis, Sujay Sanghavi
Dohyung Park, Constantine Caramanis, Sujay Sanghavi
[number, random, set, existing, median] [algorithm, max, conference, general, iid, guarantee, probability] [correct, neighborhood, ieee, computer, face, motion, pattern, based, proposed, performance, segmentation, recognition, performs, vision, table, dataset, illumination] [data, model, log, true, figure, generated, fully, standard] [subspace, clustering, point, exact, nsn, recovery, greedy, analysis, dimension, matrix, lying, lrr, ssc, ambient, condition, spectral, tsc, sparse, kmax, orthogonal, compare, closest, spanned, provided, norm, denotes, randomly, close, nuclear, basis, corresponding, gsr, pursuit] [time, wij] [nearest, computational, statistical, neighbor, theoretical, linear, problem, guaranteed, journal, error, form, machine]
Graph Clustering With Missing Data: Convex Algorithms and Analysis
Ramya Korlakai Vinayak, Samet Oymak, Babak Hassibi
Ramya Korlakai Vinayak, Samet Oymak, Babak Hassibi
[program, graph, set, edge, dmin, region, partial, consider, unweighted, number, random, partially, disjoint, identifying, constraint, red] [probability, algorithm, minimum, total, conference, explicit, transition] [work, task, output, image, computer, ieee] [data, model, figure, missing, observation, parameter, international] [matrix, adjacency, success, clustering, inside, failure, improved, nmin, analyze, recovered, recovery, succ, sbm, running, sparse, exact, compare, sujay, recovering, planted, aobs, norm, small, breed] [cluster, observed] [convex, simple, size, regularization, min, theorem, problem, note, stochastic, block, function, theoretical, ideal]
Fast Sampling-Based Inference in Balanced Neuronal Networks
Guillaume Hennequin, Laurence Aitchison, Mate Lengyel
Guillaume Hennequin, Laurence Aitchison, Mate Lengyel
[random, solution, fact, balanced, variable, provide, law] [optimal, order, appropriate, max, main] [network, learning, input, scale] [sampling, distribution, posterior, covariance, gaussian, model, mixing, inference, latent, gibbs, sampler, monte, markov, carlo, stationary, chain, sample, figure, university] [matrix, fast, faster, symmetric, diagonal, pairwise, small] [langevin, time, activity, neural, speed, recurrent, optimized, slow, noise, slowing, inverse, choice, synaptic, magnitude, weight, brain, circuit, fastest, neuronal, excitatory, state, behavior, connectivity, inhibitory, drawn, external, neuron, slowest, nonlinear, control] [stochastic, linear, large, note, machine, constant, size, statistical, cost, depends, optimization, theory, increasing]
Inferring sparse representations of continuous signals with continuous orthogonal matching pursuit
Karin C. Knudson, Jacob Yates, Alexander Huk, Jonathan W. Pillow
Karin C. Knudson, Jacob Yates, Alexander Huk, Jonathan W. Pillow
[continuous, set, discrete, constraint, sum, linearly, number] [chosen, best, choosing, seek, closely, algorithm, space] [matching, false, accurate, ieee, combination, performance] [estimate, shift, step, average, true, figure, university] [basis, cbp, waveform, comp, pursuit, svd, recovery, taylor, polar, orthogonal, greedy, dictionary, sparse, recovered, shifted, corresponding, greedily, matrix, introduced, recover, approximation, second, fourier, singular, copy, interpolate, add, small, residual, scaled, compare, distance, trace] [time, signal, neural, amplitude, spike, bin, choice, noise, estimated, width, represent, neuron, simulated] [method, convex, error, optimization, problem, update, linear, function, size, updating]
Low-dimensional models of neural population activity in sensory cortical circuits
Evan W. Archer, Urs Koster, Jonathan W. Pillow, Jakob H. Macke
Evan W. Archer, Urs Koster, Jonathan W. Pillow, Jakob H. Macke
[number, experiment] [total] [input, visual, multiple, performance, complex, well, representation, captured, cat, natural, feature] [model, latent, data, inference, true, figure, posterior, distribution, gaussian, standard, log] [subspace, matrix, small, recovered, compact, vector] [stimulus, population, neural, qlds, noise, receptive, dynamical, activity, time, cortical, nonlinear, shared, spike, system, recording, cell, poisson, correlation, single, state, simulated, capture, multiplicative, selectivity, response, sensory, drive, cortex, three, share, fit, spiking, neuron, presented, glm, distinct, represents] [quadratic, linear, function, statistical, rate, dependence, size, simple, computational, method]
Inferring synaptic conductances from spike trains with a biophysically inspired point process model
Kenneth W. Latimer, E. J. Chichilnisky, Fred Rieke, Jonathan W. Pillow
Kenneth W. Latimer, E. J. Chichilnisky, Fred Rieke, Jonathan W. Pillow
[potential, maximum, linearly, mapping, accurately, set] [total, history] [input, compared, test, output, cascade, performance, contrast, train, work, training, well] [model, likelihood, data, standard, prediction, allows, process, figure, true, log, equation, scaling, university] [point] [spike, neural, glm, excitatory, inhibitory, stimulus, conductance, cbsm, synaptic, spiking, membrane, estimated, nonlinear, fit, cell, recorded, time, include, interpretation, measured, response, simulated, intracellular, real, noise, parasol, gtot, filter, tuning, inhibition, btot, sensory, encoding, directly, retinal, exhibit, voltage, excitation, initialized, poisson, kleak, weight, biophysically, opposite, ganglion, neuron] [linear, function, rate, generalized, statistical]
Simple MAP Inference via Low-Rank Relaxations
Roy Frostig, Sida Wang, Percy S. Liang, Christopher D. Manning
Roy Frostig, Sida Wang, Percy S. Liang, Christopher D. Manning
[lrpk, map, rounding, sdp, ratio, relaxation, iqp, solution, random, relaxed, mrf, sweep, edge, number, factor, tightening, set, binary, maximum, program, marginal, rounded, interface, consider, high, polytope, integer, discrete, def, mrfs] [budget, bound, algorithm, optimal, max, lower, result, benchmark, appendix] [unit, based, table, well, dense, natural] [inference, gibbs, parallel, figure, sampling, step, chain, average, dbn, processing, advantage] [rank, matrix, pairwise, approximation, row, small, low, comparison, vector, iterative] [weight, state, underlying] [objective, problem, gradient, simple, update, theorem, quadratic, function, theoretical, randomized, optimization, constant, optimizing, solving, positive, size, stochastic, large, linear, machine]
Local Linear Convergence of Forward--Backward under Partial Smoothness
Jingwei Liang, Jalal Fadili, Gabriel Peyré
Jingwei Liang, Jalal Fadili, Gabriel Peyré
[partial, continuous, set, exists, number] [case, algorithm, total, proof, lemma, result, variation, space] [manifold, predicted, work] [local, model, restricted, tangent, figure, series, scheme] [norm, nuclear, matrix, operator, siam, proximity, sparsity, denote, subspace, vector, recovery, singular, analysis, fused, spectral] [observed, typical] [partly, convex, smooth, linear, convergence, active, theorem, relative, smoothness, journal, min, theoretical, closed, locally, rate, example, including, assumption, lasso, practical, ptx, convexity, large, regularization, gradient, converges, class, suppose, pssx, psx, pslx, function, constant, strong, machine, statistical, polyhedral, psax, addition, optimization, solve, subclass, argmin, sign, method, suppb, lipschitz, royal]
Capturing Semantically Meaningful Word Dependencies with an Admixture of Poisson MRFs
David Inouye, Pradeep K. Ravikumar, Inderjit S. Dhillon
David Inouye, Pradeep K. Ravikumar, Inderjit S. Dhillon
[number, edge, set, mrfs, high, needed, notion, easily, ordering, graph, list] [algorithm, previous, best, appendix, interesting] [word, semantic, dataset, trained, based, human, evaluation, better, training, score, work, split, similarity, idea, explicitly, well] [model, parameter, evaluate, data, approximate, parallel, modeling, likelihood, perform] [top, small, alternating, coherence, rank, speedup, component, introduced, matrix] [topic, apm, evocation, lda, admixture, poisson, semantically, meaningful, bnc, capture, capturing, directly, independent, corpus, timing, hdp, pmrf, inouye, document, quantitative, subproblems, wikipedia, qualitative] [metric, optimization, computational, simple, large, regularization, complexity, including, newton]
Dynamic Rank Factor Model for Text Streams
Shaobo Han, Lin Du, Esther Salazar, Lawrence Carin
Shaobo Han, Lin Du, Esther Salazar, Lawrence Carin
[factor, marginal, discrete, energy, number, probable, set] [associated, correlated, general, base] [word, proposed, learned, dataset, selected, based] [model, data, sampling, bayesian, latent, likelihood, distribution, gaussian, ordinal, posterior, multivariate, parameter, figure, mcmc, inference, series, modeling, monte, carlo, prior, highly, developed, temperature, covariance, step, backward] [analysis, rank, matrix, union, extended, denotes, top, sparse, united] [topic, dynamic, time, state, document, correlation, science, indicate, copula, temporal, american, admixture, poisson, country, island, tre] [framework, positive, negative, machine, linear, normal, assume, note]
A provable SVD-based algorithm for learning topics in dominant admixture corpus
Trapit Bansal, Chiranjib Bhattacharyya, Ravindran Kannan
Trapit Bansal, Chiranjib Bhattacharyya, Ravindran Kannan
[number, high, set, collection, random, paper, partition] [algorithm, probability, pure, best, satisfy, highest, called] [word, table, better, datasets, based, learn, natural, semantic, combination, multiple] [distribution, average, model, data, step, latent, full, figure] [dominant, svd, recover, clustering, provable, matrix, approximation, provably, recovery, anchor, condition, thresholding, small, weaker] [topic, document, tsvd, real, frequency, corpus, aij, dirichlet, mil, admixture, drawn, fraction, nyt, perplexity, weight, lda, vocabulary, picked, length, primary, pubmed, find] [assumption, error, note, group, problem, empirical, thresholded, assume, prove, size, dependence, pick, empirically, threshold]
Learning a Concept Hierarchy from Multi-labeled Documents
Viet-An Nguyen, Jordan L. Boyd-Graber, Philip Resnik, Jonathan Chang
Viet-An Nguyen, Jordan L. Boyd-Graber, Philip Resnik, Jonathan Chang
[node, label, set, hierarchy, number, parent, tree, political, candidate, root, structure, concept, graph, path, subtree, discover, token, ratio, edge, assigned, web] [probability, draw, associated, focus] [word, learned, test, labeled, text, improve, proposed, computed, background, additional, work, supervised, training, language, extracted, performance, unsupervised, joint, learning] [model, figure, distribution, international, sample, generated, latent, data, sampling, prediction, prior, estimate, proposal, child] [vector, compute] [topic, document, lda, hierarchical, idf, tagged, perplexity, dirichlet, tfidf, ten, taxonomy, drawn, predicting, interpretable, vocabulary, indicator, military, capture] []
On a Theory of Nonparametric Pairwise Similarity for Clustering: Connecting Clustering to Classification
Yingzhen Yang, Feng Liang, Shuicheng Yan, Zhangyang Wang, Thomas S. Huang
Yingzhen Yang, Feng Liang, Shuicheng Yan, Zhangyang Wang, Thomas S. Huang
[labeling, number, sum, marginal, exists, induced, set, equal, partition, consider] [bound, lemma, probability, associated, bandwidth, bounded, uniform, algorithm, minimum, optimal, chosen] [similarity, unsupervised, discriminative, based, learned, learning, produce, propose, correct, training, learns] [density, kernel, data, nonparametric, model, log, volume, regression, parameter, distribution, probabilistic, gaussian, average] [clustering, pairwise, low, greater, ird, corresponding, required, separation, estimating] [cluster, derived, property, connection] [generalization, error, function, pxy, estimator, hypothetical, class, nhd, consistency, nearest, neighbor, theorem, weighted, assumption, generalized, method, boundary, form, excess, derive, framework, term, depends, risk, fmax, measure, bayes]
Robust Bayesian Max-Margin Clustering
Changyou Chen, Jun Zhu, Xinhua Zhang
Changyou Chen, Jun Zhu, Xinhua Zhang
[number, set, structure, existing, random, solution, variable, maximum, augmentation, nmi] [appendix, optimal, max, space, setting] [feature, well, dataset, based, word, unsupervised, learning, test, svm, table, training, performance, supervised, compared, china, representation] [bayesian, model, posterior, latent, data, gaussian, inference, process, mixture, dpmmgm, distribution, prior, nonparametric, regbayes, mmctm, parameter, bmc, sampling, dpgmm, hyperparameters, standard, true, approach, allows, ctm, figure, generative, australian, sample, likelihood, generate] [clustering, fast, vector, compare, introduce] [topic, cluster, dirichlet, document, observed, balance, wil, generalised, sensitivity] [note, margin, key, augmented, increasing, method, framework, assumption, problem, inf]
Expectation Backpropagation: Parameter-Free Training of Multilayer Neural Networks with Continuous or Discrete Weights
Daniel Soudry, Itay Hubara, Ron Meir
Daniel Soudry, Itay Hubara, Ron Meir
[binary, discrete, number, continuous, marginal, calculation, calculated, set, map, reduce, factor, probable] [algorithm, order, optimal, online, deterministic, best, appendix, minimizing] [output, learning, training, performance, activation, backpropagation, input, layer, supervised, additional, deep, test, trained, train, better, text, datasets, network, performs, architecture, achieved] [mnns, posterior, ebp, mnn, estimate, data, distribution, bayesian, expectation, perform, approximate, pass, standard, forward, approach, hardware, factorized, restricted, propagation, parameter, university, gaussian, variational, intractable] [approximation, special] [neural, weight, real, neuronal, recall, single, synaptic, directly] [bayes, update, sign, function, computational, large, loss, calculate, note, simple, constant, assume, rate, complexity]
Attentional Neural Network: Feature Selection Using Cognitive Feedback
Qian Wang, Jiaxing Zhang, Sen Song, Zheng Zhang
Qian Wang, Jiaxing Zhang, Sen Song, Zheng Zhang
[high, random, selection] [feedback, conference, search, good, result, associated] [image, performance, segmentation, feature, background, visual, wrong, learning, input, guess, based, denoising, hidden, truth, reconstruction, correct, ground, module, attention, network, digit, better, object, trained, challenging, level, training, ann, accuracy, mnist, human, recognition, deep, cluttered, boltzmann, segment, clean, ability, discriminative, china, train, work, produce, natural, false, multiple, vision] [bias, figure, model, generative, international, processing, data, lead] [iterative, noisy, vector] [cognitive, neural, system, noise, rbm, feedforward, attentional, nature] [group, iteration, machine, framework, simple, error, example, rate, class]
Grouping-Based Low-Rank Trajectory Completion and 3D Reconstruction
Katerina Fragkiadaki, Marta Salas, Pablo Arbelaez, Jitendra Malik
Katerina Fragkiadaki, Marta Salas, Pablo Arbelaez, Jitendra Malik
[complete, high, structure, red, number] [benchmark, sequence, algorithm, best, case, mild] [video, reconstruction, object, camera, motion, dense, segmentation, rigid, image, work, optical, depth, monocular, rotation, proposed, frame, occluded, input, pose, reprojection, bilinear, representation, propose] [missing, figure, incomplete, data] [rank, matrix, nrsfm, nuclear, point, norm, euclidean, synthetic, upgrade, factorization, recover, completion, denote, nonrigid, basis, subspace, minimization, clustering, compute, row, small, projection] [trajectory, shape, cluster, tracking, varying, temporal, observed, subject, real, orthonormality, realistic, perception] [method, error, regularized, linear, min, term, journal, large]
Learning Mixtures of Submodular Functions for Image Collection Summarization
Sebastian Tschiatschek, Rishabh K. Iyer, Haochen Wei, Jeff A. Bilmes
Sebastian Tschiatschek, Rishabh K. Iyer, Haochen Wei, Jeff A. Bilmes
[set, collection, scoring, quality, high, subset, number, consider, random, maximization, complement, graph, sum] [submodular, monotone, algorithm, conference, best, good, coverage, budget, worst] [image, summarization, human, summary, score, visual, learning, evaluation, computer, performance, diversity, submodularity, similarity, natural, based, vision, computed, color, proposed, work, surrogate, scene, structured, reference, learnt, collected, representative, learn, baseline, personal, rouge, ground, location, video, table] [data, average, mixture, international, approach, volume, model] [greedy, component, element, facility, clustering, low, introduce, intermediate, amazon, small] [described, document, property, weight, include] [function, loss, problem, objective, large, optimization, method, machine, optimize]
Deep Learning Face Representation by Joint Identification-Verification
Yi Sun, Yuheng Chen, Xiaogang Wang, Xiaoou Tang
Yi Sun, Yuheng Chen, Xiaogang Wang, Xiaoou Tang
[number, set, classify, reduce] [previous, best, algorithm, probability] [face, feature, learning, learned, deep, accuracy, joint, extracted, training, identity, lfw, yij, recognition, representation, compared, convolutional, supervisory, image, test, selected, extract, extraction, layer, reducing, diverse, similarity, convnet, entire, learn, scatter, report, table, well, input, dataset, diversity, achieved, trained] [bayesian, figure, model, target, log, sample, data, university] [norm, cosine, comparison, technical, pca, corresponding, subspace] [signal, time, system, chinese, distinguishing] [large, effective, increasing, metric, loss, function, linear, generalized, updated, denoted]
Fast Training of Pose Detectors in the Fourier Domain
João F. Henriques, Pedro Martins, Rui F. Caseiro, Jorge Batista
João F. Henriques, Pedro Martins, Rui F. Caseiro, Jorge Batista
[consider, solution, discrete, deal, structure] [appendix, period, case, cycle] [pose, image, training, virtual, learning, transformation, object, natural, dft, transformed, datasets, rotation, detection, circulant, train, multiple, planar, trained, work, translation, structured, walk, proposed, detector, based, computer, accelerate, recognition, svr, learn, array, table, transform, original, applied, kitti] [data, sample, model, regression, allows, estimate, demonstrate] [matrix, vector, fourier, support, orthogonal, fast, basis, operator, ridge, row, denotes, slice, compute, matlab, smaller] [single, represent, neural] [cyclic, negative, method, computational, problem, large, dual, require, estimation, linear]
PEWA: Patch-based Exponentially Weighted Aggregation for image denoising
Charles Kervrann
Charles Kervrann
[set, procedure, consider, number, collection, candidate, random] [aggregation, algorithm, best, current, considered, general, aggregated] [image, pewa, patch, denoising, denoised, ieee, psnr, input, proposed, transform, table, multiple, ewa, dct, produce, clean, improve, combined, neighborhood, performed, natural, castle, performance, based, learning, fpewa, epll, barbara] [average, gaussian, prior, exponential, implementation, markov, processing, comparable, chain, figure, standard, sampling, computation] [noisy, compute, man, sparse, second, corrupted, comparison] [noise, white, external, internal, three, single, assumed] [estimator, method, weighted, estimation, large, basic, form, risk]
A Multi-World Approach to Question Answering about Real-World Scenes based on Uncertain Input
Mateusz Malinowski, Mario Fritz
Mateusz Malinowski, Mario Fritz
[set] [type, uncertain] [semantic, human, spatial, based, image, question, language, object, table, segmentation, answer, scene, answering, visual, wups, accuracy, automatic, test, training, natural, dataset, architecture, multiple, learning, task, performance, vision, autoseg, complex, work, representation, well, chair, reasoning, propose, front, reference, computer, turing, xmean, color, zmean, humanseg] [approach, figure, latent, probabilistic, inference, data, uncertainty, generated] [synthetic, row] [single, perceived, system] [logical, measure, class, form, method]
Quantized Kernel Learning for Feature Matching
Danfeng Qin, Xuanli Chen, Matthieu Guillaumin, Luc V. Gool
Danfeng Qin, Xuanli Chen, Matthieu Guillaumin, Luc V. Gool
[number, interval, binary, set, label] [space, explicit, best, conference, design, algorithm, uniform, adaptive, base, piecewise] [feature, learning, quantization, performance, sift, matching, learn, image, descriptor, qks, table, false, similarity, quantizer, ieee, pattern, quantizers, yosemite, visual, computer, original, work, based, vision, training, multiple, learnt, higher, est, compression] [kernel, data, local, true, parameter, observation, model] [matrix, euclidean, additive, product, rank, dimension, compact, liberty, denotes, analysis] [single, recall, three] [quantized, positive, optimization, rate, machine, family, negative, block, problem, large, constant, simple, metric, convex]
Diverse Sequential Subset Selection for Supervised Video Summarization
Boqing Gong, Wei-Lun Chao, Kristen Grauman, Fei Sha
Boqing Gong, Wei-Lun Chao, Kristen Grauman, Fei Sha
[subset, selection, set, existing, quality, agreement, consider, map] [oracle, algorithm, selecting, contextual, main, probability] [video, summarization, diverse, learning, summary, visual, frame, seqdpp, vsumm, ground, select, evaluation, supervised, representation, unsupervised, novel, selected, youtube, idea, training, powerful, despite, diversity, well, human, learn, propose, alex, ben, performance, automatic, computer, kodak, datasets, work, table] [dpp, sequential, model, determinantal, modeling, process, standard, approach, data, markov, prior, application, parameter, university] [point, vector, randomly, greater, compute, pairwise] [document, temporal, science, neural] [linear, method, arg, averaged, note, key]
LSDA: Large Scale Detection through Adaptation
Judy Hoffman, Sergio Guadarrama, Eric S. Tzeng, Ronghang Hu, Jeff Donahue, Ross Girshick, Trevor Darrell, Kate Saenko
Judy Hoffman, Sergio Guadarrama, Eric S. Tzeng, Ronghang Hu, Jeff Donahue, Ross Girshick, Trevor Darrell, Kate Saenko
[set, region, map, scoring, consider] [algorithm, type, conference, search] [detection, bounding, category, network, object, adaptation, box, training, image, labeled, layer, detector, background, false, annotated, lsda, convolutional, train, imagenet, adapt, trained, learn, fcbgrnd, feature, learning, deep, visual, produce, transformation, cnn, performance, adapting, output, scale, dataset, validation, challenge, localization, task, computer, invariant, learns, oth, baseline, collect, transfer, svms, localizes] [data, figure, model, full, approach, proposal, source, international] [top, analysis, confusion, requires, technique] [neural, time, directly, held] [large, domain, positive, error, method, linear]
Local Decorrelation For Improved Pedestrian Detection
Woonhyun Nam, Piotr Dollar, Joon Hee Han
Woonhyun Nam, Piotr Dollar, Joon Hee Han
[random, code, number, tree, topology, oriented] [decision, result, correlated, observe, van] [oblique, pedestrian, decorrelation, decorrelated, channel, training, ldcf, detection, acf, feature, boosted, work, object, dataset, natural, image, learning, baseline, caltech, trained, inria, detector, original, representation, propose, false, accuracy, split, additional, considerable, depth, performance, table, well, transform, spatially, based, report, patch, capacity, improves, proposed, improvement, decorrelating, deep, validation] [local, data, covariance, model, figure, approach, gaussian, sampling, straightforward] [orthogonal, matrix, eigenvectors, fast, computing, projection, top] [lda, single, shared, window] [locally, boosting, linear, size, computational, rate, effective, increasing, function, autocorrelation, constant, large]
Recursive Inversion Models for Permutations
Christopher Meek, Marina Meila
Christopher Meek, Marina Meila
[structure, rim, permutation, number, set, truct, left, sushi, tree, consider, meath, typically, huang, node, sum, binary, subtree, guestrin, vertex, meek, anonical, ordering, recursively, polynomial, discrepancy] [algorithm, search, proposition, optimal, order, alternative, associated, best] [reference, recursive, work, learned, well, test, indicating, computed, split] [model, data, parameter, gmm, inversion, likelihood, sample, independence, estimate, approach, child, earch, dij, figure, local, distribution, international, canonical] [matrix, element, estimating, rank, denote, distance] [internal, preference, observed, time, merging, capture] [note, estimation, respect, generalized, optimization, strong, negative, machine, problem, corresponds]
QUIC & DIRTY: A Quadratic Approximation Approach for Dirty Statistical Models
Cho-Jui Hsieh, Inderjit S. Dhillon, Pradeep K. Ravikumar, Stephen Becker, Peder A. Olsen
Cho-Jui Hsieh, Inderjit S. Dhillon, Pradeep K. Ravikumar, Stephen Becker, Peder A. Olsen
[selection, consider, structure, set, sum, apply, number, graphical, variable, solution] [algorithm, current, appendix, general, proof, proposition] [dataset, learning, propose] [parameter, approach, covariance, latent, model, global, figure, gaussian] [subspace, matrix, sparse, norm, hessian, approximation, alternating, faster, minimization, support, special, vector, component] [property, inverse, free] [quadratic, proximal, loss, active, solve, subproblem, convergence, newton, gradient, descent, problem, decomposable, statistical, group, sfree, method, framework, note, convex, dirty, solving, coordinate, function, class, direction, objective, regularizers, pgalm, solved, rate, positive, regularization, structural, xgt, size, min, converges, written, logistic, key, estimation, optimization, block, form, argmin, update]
Learning Chordal Markov Networks by Dynamic Programming
Kustaa Kangas, Mikko Koivisto, Teppo Niinimäki
Kustaa Kangas, Mikko Koivisto, Teppo Niinimäki
[tree, junction, set, programming, gobnilp, chordal, rooted, junctor, rpt, consider, number, subset, clique, node, structure, recurrence, path, hypergraph, partition, characterization, subtree, maximum, vertex, requirement, graph, factor, maximizes, enables, disjoint, graphical, integer, constraint, acyclic, maximization, root, scoring, induction, random, edge, treewidth, subtrees] [algorithm, optimal, max, observe, case, bounded, space, bound, benchmark, intersection, search, call] [learning, score, network, recursive, work, based, performance, evaluation] [markov, bayesian, data, figure, observation, approach] [running, product, equivalent, denote, synthetic, small] [dynamic, time, three, width, reduced] [assume, basic, machine, problem, size, solve, journal, theorem, function, linear]
Optimization Methods for Sparse Pseudo-Likelihood Graphical Model Selection
Sang Oh, Onkar Dalal, Kshitij Khare, Bala Rajaratnam
Sang Oh, Onkar Dalal, Kshitij Khare, Bala Rajaratnam
[graphical, partial, selection, number, high, graph, set, cancer] [algorithm, initial, bound, lower, upper, literature] [proposed, table, datasets, dataset, based] [log, step, gaussian, covariance, approach, model, breast, likelihood, scalable, demonstrate, sample, data, national] [sparse, matrix, diagonal, synthetic, algebra, backtracking, numerical, minimization, iterative, second, running, analysis] [inverse, correlation, real, fastest] [concord, size, ccista, estimation, convergence, gradient, proximal, qqqq, ccfista, constant, method, qcon, iter, function, convex, framework, problem, linear, objective, optimization, computational, det, term, iteration, tremendous, theorem, smooth, journal, address, prove, lipschitz, complexity, ista, accelerated, dimensional]
Graphical Models for Recovering Probabilistic and Causal Queries from Missing Data
Karthika Mohan, Judea Pearl
Karthika Mohan, Judea Pearl
[missingness, causal, recoverable, recoverability, set, attrition, mohan, variable, path, graphical, mechanism, collider, substantive, partially, pearl, connected, asub, corollary, admissible, notion, graph, longitudinal, exists, complete, der, shorthand, apply, mar, laan, parent, manifest, minimal, rvi, marlin, permit] [general, called, appendix, sequence, algorithm, case] [joint, based, dataset, work] [missing, data, figure, probabilistic, distribution, conditional, model, sequential, imputation, latent, inducing] [recovering, factorization, condition, ordered, recovery, intermediate, corrupted] [observed, presented, common] [theorem, example, query, consistent, statistical, form, simple, journal, class, discussed, hold]
Divide-and-Conquer Learning by Anchoring a Conical Hull
Tianyi Zhou, Jeff A. Bilmes, Carlos Guestrin
Tianyi Zhou, Jeff A. Bilmes, Carlos Guestrin
[number, set, variable, subset, random, reduce, solution, covered, unique, minimal, maximum] [general, minimum, algorithm, order, probability, proposition, case] [learning, accuracy, original, hyperplane, multiple, table, hidden, training, motion] [latent, data, model, figure, gmm, mixture, likelihood, sampling, markov, log, sample, separable, hmm, gibbs] [dca, conical, hull, matrix, spectral, separability, clustering, factorization, anchor, cpu, point, subspace, nmf, anchoring, fast, comparison, angle, cone, analysis, projection, spa, randomly, xray, sfo, column] [time, experimental, lda, real, dirichlet] [problem, method, assumption, solving, large, form, solver, convex, acc, solved]
Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimation
Jonathan J. Tompson, Arjun Jain, Yann LeCun, Christoph Bregler
Jonathan J. Tompson, Arjun Jain, Yann LeCun, Christoph Bregler
[set, number, unary, existing, overlapping, graph, structure] [lower] [body, joint, training, relu, conv, human, pose, network, convolution, fig, performance, input, shoulder, image, flic, detection, output, face, softplus, resolution, convolutional, architecture, spatial, wrist, bank, learning, dataset, pool, toshev, elbow, test, work, lsp, pixel, evaluation, articulated, trained, feature, pishchulin, scene, sliding, ankle, improve, improves, location, knee, deep, convnet, sapp, predicted, discriminative, learn] [model, figure, log, full, prior, distribution] [distance, equivalent, stage] [single, receptive, create, window] [estimation, normalized, error, large, rate, problem]
A Complete Variational Tracker
Ryan D. Turner, Steven Bottone, Bhargav Avasarala
Ryan D. Turner, Steven Bottone, Bhargav Avasarala
[assignment, map, belief, number, bethe, entropy, paper, induced] [algorithm, lower, bound, associated, probability, player, space] [accuracy, association, multiple, sliding, computer, vision, frame, ieee] [track, clutter, data, variational, mht, tracker, posterior, log, prior, model, figure, measurement, approach, full, ari, radar, conjugate, omgp, process, soccer, inference, standard, propagation, develop, management, framing, approximate, distribution, markov, cap, probabilistic, spurious, kalman, exponential, processing, missed, estimate, siap] [matrix, row, compute, vector] [tracking, state, window, aij, loopy, real, time, free] [problem, example, linear, method, update, cost, note, loss, form, active]
Concavity of reweighted Kikuchi approximation
Po-Ling Loh, Andre Wibisono
Po-Ling Loh, Andre Wibisono
[bethe, kikuchi, region, reweighted, graph, concavity, entropy, partition, belief, set, attractive, sum, concave, wainwright, polytope, message, mixed, contained, graphical, ising, unweighted, potential, passing, structure, provide, forest, edge, characterization, pakzad, strictly, consider, msr, spanning, collection, anantharam, energy] [algorithm, case, appendix, result, general, proof, version, main, upper, interesting] [multiple, representation, ieee] [log, local, propagation, variational, distribution, model, global, equation, processing, figure] [approximation, product, condition, pairwise, establish, corresponding, denote, point] [loopy, neural, free, weight] [function, theorem, objective, note, problem, generalized, convergence, theory]
Sequential Monte Carlo for Graphical Models
Christian Andersson Naesseth, Fredrik Lindsten, Thomas B. Schön
Christian Andersson Naesseth, Fredrik Lindsten, Thomas B. Schön
[graphical, partition, set, number, random, factor, graph, pgm, partial, consider, belief, pgms, undirected, annealing, ordering, construct, tree, structure, collection] [algorithm, sequence, probability, conference, general, interesting, choose, case] [proposed, joint, learning, compared, performance] [smc, sampler, particle, sequential, pmcmc, target, figure, monte, model, inference, carlo, latent, distribution, mcmc, gibbs, markov, del, asir, sample, standard, auxiliary, blocking, hamze, moral, data, estimate, full, probabilistic, kernel, freitas, density, xlk, conditional, unbiased, ancestor] [decomposition, corresponding, intermediate, exact] [topic] [method, statistical, function, journal, simple, framework, estimator, computational, example, machine]
Distributed Estimation, Information Loss and Exponential Families
Qiang Liu, Alex T. Ihler
Qiang Liu, Alex T. Ihler
[consider, random, arbitrary, number, john, partition] [total, lower, general, minimum, order, study, bound, setting, best] [learning, combination, training, fisher, based, test, natural, performs] [local, global, exponential, data, sample, average, mle, full, model, curvature, parameter, true, gaussian, curved, distribution, mixture, averaging, log, efron, likelihood, partitioned, merugu, generative, toy, summarizing, figure] [square, exactly, randomly, exact, suggests, arxiv, symmetric] [represents] [linear, distributed, size, statistical, family, loss, mles, method, asymptotic, empirical, theorem, theoretical, example, error, consistent, communication, large, theory, note, practical, function, assume, rate, machine, geometric]
Robust Kernel Density Estimation by Scaling and Projection in Hilbert Space
Robert A. Vandermeulen, Clayton Scott
Robert A. Vandermeulen, Clayton Scott
[set, construct, robustness, high, exists, label, asymptotically] [version, anomaly, max, proposition, uniform, proof, algorithm, difference, bandwidth, satisfy, lemma, result] [detection, level, well, performance, transform, datasets, test, transformed, applied, performs, training, traditional, testing] [density, ftar, spkde, fcon, contamination, kde, kernel, sample, dkl, nonparametric, figure, target, pdf, rkde, data, estimate, dcon, decontaminates, approach, model, contaminating, log, parameter, scott, wilcoxon, parametric, lebesgue, supplemental, scaling, gaussian, kim] [robust, scaled, recover, support, rank, component, projection, hilbert, contaminated, introduce] [investigate, experimental] [assumption, estimation, method, estimator, divergence, asymptotic, converges, consistency, metric, theorem, quadratic, weighted, class, threshold, convex, closed]
SerialRank: Spectral Ranking using Seriation
Fajwel Fogel, Alexandre d'Aspremont, Milan Vojnovic
Fajwel Fogel, Alexandre d'Aspremont, Milan Vojnovic
[number, set, permutation, ordering, item] [ranking, algorithm, ranked, order, case, proposition, study, total, minimum] [similarity, based, produce, pair, identity, score, learning] [true, missing, figure, model, data] [pairwise, matrix, seriation, spectral, corrupted, match, man, west, strict, serialrank, fiedler, villa, aston, kendall, vector, top, synthetic, comparison, rank, point, united, arsenal, norwich, hull, city, palace, recover, southampton, crystal, swansea, brom, ham, fulham, newcastle, everton, sunderland, noisy, cardiff, tottenham, link, compute, chelsea, ordered, liverpool, stoke, reorder, vary, recovers, consists, condition, recovery] [glm, observed, real, classical] [problem, method, linear, consistent, generalized, increasing, suppose]
Top Rank Optimization in Linear Time
Nan Li, Rong Jin, Zhi-Hua Zhou
Nan Li, Rong Jin, Zhi-Hua Zhou
[number, bipartite, solution, existing, binary, root, high, set, truncated] [ranking, ranked, algorithm, probability, optimal, max, version, result, maximize] [training, accuracy, learning, performance, based, table, proposed, evaluation, learn, datasets, well] [step, data, conjugate, log, develop, developed, approach] [top, projection, leading, analysis, rank, pairwise, vector, support, largest] [time, auc, include] [toppush, positive, loss, negative, function, linear, instance, computational, dual, complexity, problem, large, method, svmpauc, convex, svmrank, optimization, optimizing, theoretical, theorem, objective, key, empirical, svmmap, convergence, aatp, gradient, optimize, cost, address, size, rate, min, logistic, solve]
Factoring Variations in Natural Images with Deep Gaussian Mixture Models
Aaron van den Oord, Benjamin Schrauwen
Aaron van den Oord, Benjamin Schrauwen
[number, path, maximization, factor, set, paper, construct, node] [heuristic, algorithm, current, conference, case, best, easy, called] [deep, learning, layer, image, transformation, training, unsupervised, natural, multiple, shallow, dataset, better, pixel, train, performance, tiny, network, vec, trained, propose, yoshua, transformed] [gmm, mixture, model, log, gmms, data, gaussian, density, figure, international, equation, parameter, generative, scalable, standard, expectation, distribution, processing, likelihood, average, rnade, modeling] [top, initialization] [neural, represent, optimized] [machine, optimization, gradient, normal, large, method, optimizing, estimation, objective, function, constant, stochastic]
Near-Optimal Density Estimation in Near-Linear Time Using Variable-Width Histograms
Siu On Chan, Ilias Diakonikolas, Rocco A. Servedio, Xiaorui Sun
Siu On Chan, Ilias Diakonikolas, Rocco A. Servedio, Xiaorui Sun
[set, partition, interval, discrete, paper, high, number] [algorithm, probability, optk, unknown, dtv, lemma, optimal, optc, main, access, piecewise, goal, consecutive, case, variation, total, appendix, breakpoint, result, proof, bound, general] [learning, input, output, accuracy, question] [distribution, sample, density, target, step, pdf, model, parameter, standard, approach, data, approximate] [distance, running, denote, approximation, dimension, reduction, add] [time, independent, drawn, weight, rest] [theorem, hypothesis, size, estimation, constant, problem, statistical, empirical, linear, prove, domain, note, agnostic, function, inequality, theory, class, large, complexity, including, assumption, simple, unimodal, algorithmic]
Learning Mixed Multinomial Logit Model from Ordinal Data
Sewoong Oh, Devavrat Shah
Sewoong Oh, Devavrat Shah
[mixed, number, partial, graph, random, permutation, high, provide, discrete, exists, dmin] [algorithm, probability, chosen, associated, result, type, bounded, established, order, choose, phase, ranking, ranked, max, main, transition] [learning, learn, long, work, based, well, learnt, score] [mixture, model, distribution, data, log, estimate, sample, chain, logit, markov, latent, computationally, observation, approach] [mnl, matrix, tensor, spectral, component, dmax, rank, denote, ank, decomposition, qmin, second, pka, completion, incoherence, entrality, diagonal, qmax, introduced, condition, approximation] [choice, recall, preference, observed, individual, state, statistically] [theorem, estimation, error, positive, method, assume, large, example, machine, problem]
Magnitude-sensitive preference formation`
Nisheeth Srivastava, Ed Vul, Paul R. Schrater
Nisheeth Srivastava, Ed Vul, Paul R. Schrater
[set, label, item, hill, making, forest, correspond, high, rational, ordering] [price, probability, economic, utility, context, option, history, demand, decision, expected, pareto, formation, general, pricing, evidence, receiving] [learning, human] [model, figure, distribution, equation, data, larger, observation, latent, sampling] [absolute] [choice, preference, money, monetary, risky, wealth, psychophysical, behavior, desirability, directly, classical, panel, involving, typical, wine, psychology, preferred, giffen, contribution, three, encodes, presented, shock, orbitofrontal, experience, animal, brain, frequency, inferred] [theory, relative, term, journal, large, rate, framework, form, assume, risk, simple]
Robust Classification Under Sample Selection Bias
Anqi Liu, Brian Ziebart
Anqi Liu, Brian Ziebart
[reweighted, selection, label, maximum, random, entropy, high, set, uci, solution] [minimax, minimizes, conference, minimizing, chosen] [learning, dataset, feature, performance, based, input, matching, datasets, despite] [target, source, distribution, sample, approach, data, conditional, bias, psrc, logloss, prediction, density, ptrg, figure, model, epsrc, regression, likelihood, eptrg, covariate, rba, probabilistic, shift, inductive, uncertainty, estimate, moment, international, processing, log, datapoints, parametric, reweighting] [robust, knowledge, minimization, support, generalize, compare] [neural, estimated, measured, certainty, assumed] [estimation, logistic, loss, machine, large, strong, generalization, optimization, convex, statistical, regularization, gradient, theorem, function, problem, weighting]
An Accelerated Proximal Coordinate Gradient Method
Qihang Lin, Zhaosong Lu, Lin Xiao
Qihang Lin, Zhaosong Lu, Lin Xiao
[set, random, number, composite, apply, def, partial] [algorithm, choose, case, minimize, general, minimizing, order, result, exploit] [learning, report] [parameter, full, implementation, developed, international] [compute, vector, avoid, repeat, uniformly, siam, norm, lee, technical] [ascent] [method, coordinate, apcg, gradient, accelerated, descent, convergence, convex, block, convexity, sdca, dual, function, iteration, proximal, regularized, linear, stochastic, erm, complexity, solving, randomized, assumption, optimization, theorem, strong, machine, min, dom, problem, uik, journal, update, vik, afg, objective, arg, primal, cost, respect, rni, loss, updated, risk, regularization, atk, note, empirical, solve, rate, theory, lipschitz]
Stochastic Gradient Descent, Weighted Sampling, and the Randomized Kaczmarz algorithm
Deanna Needell, Rachel Ward, Nati Srebro
Deanna Needell, Rachel Ward, Nati Srebro
[consider, number, proportional, partially, solution, factor] [bound, uniform, probability, case, guarantee, minimizer, chosen, considered] [improve, based, improvement, applied, better, performance, accuracy] [sampling, distribution, average, variance, source, sample, spherical, allows, figure, unbiased, fully, gaussian, standard, mixture, expectation, exponential] [analysis, denote, second, residual, matrix, row] [desired, weight, drawn] [sgd, linear, dependence, convergence, method, kaczmarz, weighted, convex, lipschitz, randomized, gradient, stochastic, iteration, theorem, biased, quadratic, weighting, objective, error, smooth, constant, conditioning, bach, iterates, quantity, optimization, function, term, depends, smoothness, complexity, solving, squared, overdetermined, respect, supremum, moulines, strohmer]
Rounding-based Moves for Metric Labeling
M. Pawan Kumar
M. Pawan Kumar
[rounding, labeling, random, interval, complete, energy, label, unary, procedure, solution, variable, factor, assigned, wab, relaxation, set, edge, arbitrary, graph, number, programming, feasible, equal, truncated, tight, overestimation, integer, unique, yab, consider, dxa, assign, mrf, fractional, map, labelings, constraint] [algorithm, submodular, bound, minimum, current, probability, design, order, main, general, call, optimal] [level, proposed, input, computed] [parallel, approximate, approach, markov, step] [distance, approximation, consists, special, pairwise, clustering, minimization, corresponding, compute] [hierarchical, cluster, multiplicative, single, range] [metric, function, problem, linear, solving, theoretical, optimization, convex, note, theorem, large, inequality, randomized, distortion, assigning, expansion]
Two-Stream Convolutional Networks for Action Recognition in Videos
Karen Simonyan, Andrew Zisserman
Karen Simonyan, Andrew Zisserman
[number, fusion, consider] [action, consecutive] [training, optical, convnet, spatial, video, recognition, input, displacement, motion, architecture, deep, learning, convolutional, stream, network, trained, stacking, image, layer, frame, stride, based, computed, dense, accuracy, softmax, table, additional, multiple, dropout, human, dataset, convnets, evaluation, shallow, better, performance, net, challenge, performs, stacked, split, pool, scratch, visual, feature, horizontal, representation, hmax, learnt, combined, stack] [data, local, volume, model, averaging, implementation] [point, vector] [temporal, trajectory, single, state, described, individual, neural] [large]
Semi-supervised Learning with Deep Generative Models
Diederik P. Kingma, Shakir Mohamed, Danilo Jimenez Rezende, Max Welling
Diederik P. Kingma, Shakir Mohamed, Danilo Jimenez Rezende, Max Welling
[number, label, set, variable, construct, existing, graph] [bound, conference, algorithm, best, space, exploit, lower] [learning, deep, performance, training, discriminative, mnist, image, based, feature, hidden, unsupervised, manifold, test, learn, network, embeddings, combined, accurate, competitive, combine, table] [model, data, generative, inference, latent, variational, distribution, log, gaussian, posterior, approach, allow, approximate, international, likelihood, conditional, processing, demonstrate, optimisation, style, svhn, tsvm, probabilistic, scalable, perform, allows, analogical, density] [corresponding, transductive] [neural, range] [class, labelled, stochastic, objective, unlabelled, machine, form, cost, complexity, function, large, simple]
Searching for Higgs Boson Decay Modes with Deep Learning
Peter J. Sadowski, Daniel Whiteson, Pierre Baldi
Peter J. Sadowski, Daniel Whiteson, Pierre Baldi
[set, complete, discovery, collider, increase, number, provide] [difference, best, initial] [deep, learning, network, shallow, hidden, trained, training, table, background, learn, performance, computer, invariant, feature, better, complex] [data, figure, mass, particle, hyperparameter, process, hyperparameters, missing, demonstrate, distribution, university, standard, approach, prediction] [analysis, power, intermediate, technique] [neural, fraction, higgs, boson, momentum, lepton, decay, state, observed, simulated, signal, raw, hadron, auc, standardized, stable, collision, derived, lhc, issn, transverse, experimental, irvine] [machine, large, rate, statistical, direction]
Learning to Discover Efficient Mathematical Identities
Wojciech Zaremba, Karol Kurach, Rob Fergus
Wojciech Zaremba, Karol Kurach, Rob Fergus
[grammar, expression, symbolic, degree, random, number, mathematical, tree, set, naive, continuous, paper, aat, simpler, solution, discover, discovery, scheduler, consider, apply, grows, operation, multiply, valid, binary] [search, strategy, space, current, lower] [learning, representation, learn, recursive, rnn, training, complex, softmax, attribute, evaluation, learned, language, discovered, table, network, work, descriptor, combination, applied, output, predict, based, novel, identity, well, depth] [target, explore, exponential, model, computation, allows, figure] [vector, matrix, involves, numerical, match, repeat] [neural, rule, single, weight, system] [machine, linear, framework, complexity, simple, class, example, problem, large, computational]
Asynchronous Anytime Sequential Monte Carlo
Brooks Paige, Frank Wood, Arnaud Doucet, Yee Whye Teh
Brooks Paige, Frank Wood, Arnaud Doucet, Yee Whye Teh
[number, marginal, set, queue, random, programming, synchronization] [algorithm, initial, total, order, branching, space, probability, proposition] [cascade, additional, performance, produce, frank] [particle, resampling, smc, monte, sequential, carlo, likelihood, process, anytime, child, posterior, observation, average, forward, markov, icsmc, estimate, approach, distribution, inference, blocking, probabilistic, gaussian, chain, mse, arrive, parallel, fully, deterministically, arnaud, scheme, latent, implementation, interacting, hardware, limit, conditional, pierre, step, simulation, true, international, computation] [running, computing, compute, faster] [weight, system, memory, single, state, time, count, simultaneously, independent, choice, estimated] [large, convergence, simple, linear, statistical, note, suppose, asynchronous, iterated]
Probabilistic ODE Solvers with Runge-Kutta Means
Michael Schober, David K. Duvenaud, Philipp Hennig
Michael Schober, David K. Duvenaud, Philipp Hennig
[solution, differential, structure, constructed, marginal, high, construct, remaining] [order, probability, algorithm, good, initial, return, open, previous] [evaluation, scale, question, based, output, table, matching] [posterior, process, probabilistic, wiener, gaussian, prior, extrapolation, hci, distribution, integrated, kernel, covariance, limit, third, parameter, gmrk, true, hennig, figure, standard, regression, global, inference, series, observation, analogous, prediction, conceptual, variance, bayesian, hauberg, butcher, model] [second, match, numerical, point, continuation] [choice, three, interpretation, rise, observed, integration] [ode, method, linear, family, function, ordinary, convergence, machine, gradient, proper, theoretical, solver, smoothing, argument, error, consistent, problem, estimation, theorem]
A Wild Bootstrap for Degenerate Kernel Tests
Kacper P. Chwialkowski, Dino Sejdinovic, Arthur Gretton
Kacper P. Chwialkowski, Dino Sejdinovic, Arthur Gretton
[procedure, random, core, number, construct, set, sum, degree] [type, probability, alternative, proposition, bounded, lemma, main, space, case, called, version, write, limiting] [test, testing, based, performance, multiple, embeddings, embedding, proposed, applied] [bootstrap, kernel, null, distribution, wild, independence, process, bootstrapped, sample, series, hsic, degenerate, mmd, gibbs, stationary, statistic, kcsd, mcmc, converge, approach, volume, shift, zim, figure, auxiliary, behaviour, parameter] [hilbert, second] [time, independent, dependent, temporal, instantaneous, panel, audio, desired] [error, dependence, hypothesis, empirical, asymptotic, assumption, theorem, function, lipschitz, statistical, converges, consistent, rate, convergence, large, method, normalized, consistency]
(Almost) No Label No Cry
Giorgio Patrini, Richard Nock, Tiberio Caetano, Paul Rivera
Giorgio Patrini, Richard Nock, Tiberio Caetano, Paul Rivera
[label, map, set, scoring, proportion, consider, entropy, assignment, minimal, random, solution, provide, binary, exists] [algorithm, lemma, setting, best, oracle, unknown, uniform] [learning, bag, feature, supervised, table, manifold, surrogate, better, learns] [step, big, statistic, true, figure, data, fully] [operator, small, matrix, denote, approximation, iterative, laplacian, provided, square, corresponding] [auc] [amm, lmm, loss, llp, vjj, assumption, homogeneity, ammmin, linear, subsection, problem, estimation, generalization, logistic, display, function, convergence, risk, consistent, complexity, rademacher, arg, respect, theorem, proper, convex, large, emm, empirical, estimator, invcal, theoretical, hold, machine, relative, instance, optimizes, assoc, min]
Consistent Binary Classification with Generalized Performance Metrics
Oluwasanmi O. Koyejo, Nagarajan Natarajan, Pradeep K. Ravikumar, Inderjit S. Dhillon
Oluwasanmi O. Koyejo, Nagarajan Natarajan, Pradeep K. Ravikumar, Inderjit S. Dhillon
[binary, consider, fundamental, set, asymmetric] [optimal, algorithm, probability, result, proposition, consisting, benchmark, conference, utility, depend, appendix, decision, setting, lemma] [performance, training, surrogate, accuracy, proposed, false, dataset, learning, work, similarity, table, computed] [estimate, data, international, conditional, true, standard, distribution, university, sample] [analysis, special] [population, presented] [threshold, measure, loss, family, class, positive, consistent, weighted, bayes, metric, risk, machine, empirical, theorem, function, logistic, depends, sign, erm, statistical, linear, simple, respect, consistency, journal, thresholded, large, theoretical, rate, note, generalized, cene, address, optimizing, practice, characterized, assumption]
Extended and Unscented Gaussian Processes
Daniel M. Steinberg, Edwin V. Bonilla
Daniel M. Steinberg, Edwin V. Bonilla
[experiment, number, solution] [lower, algorithm, bound, order, kak, setting] [learning, test, well, performance, table, recursive, based, cnn] [likelihood, log, posterior, gaussian, variational, ugp, linearization, egp, latent, inference, unscented, forward, process, model, equation, inversion, kalman, linearized, series, expectation, kernel, infer, jmk, evaluate, predictive, gps, prior, figure, data, hyperparameters, intractable, regression, processing, step, kinematics, prediction] [approximation, extended, taylor, iterative, trace, factorizing] [nonlinear, inverse, system, dynamical, presented, neural] [function, statistical, method, machine, term, update, error, derive, objective, iterated, linearizing, framework, require, logistic, dimensional, computational, negative, rate]
Controlling privacy in recommender systems
Yu Xin, Tommi Jaakkola
Yu Xin, Tommi Jaakkola
[privacy, private, public, number, rating, user, mechanism, set, marginal, recommender, item, release, limited, random, differential, discrete, consider, maintaining, concept, solution, making, lap] [order, probability, bound, algorithm, access, total, controlled, lemma, case, optimal] [accuracy, test, performance, based, additional, select] [distribution, rmse, full, log, prediction, parameter, demonstrate, figure, restricted, estimate, data] [matrix, vector, second, norm, trace, low, exact, rank, completion, denote, small, compare, preserving, element, subspace, analysis, estimating] [noise, observed, underlying, share, system] [server, assume, regularization, note, depends, error, method, estimation, weighted, function, statistical, large, problem, theorem, strong, solving]
Probabilistic low-rank matrix completion on finite alphabets
Jean Lafond, Olga Klopp, Eric Moulines, Joseph Salmon
Jean Lafond, Olga Klopp, Eric Moulines, Joseph Salmon
[consider, number, set, recommender, maximum, random, high, rating, partial, map, paper] [algorithm, case, probability, associated, general, upper, bound, unknown, version, considered, max, space] [table, proposed, learning] [gaussian, prediction, distribution, sampling, model, parameter, true, data, logit, sampled, log, computation, allows] [matrix, completion, singular, rank, vector, norm, nuclear, denote, requires, minimization, noisy, svd, tensor, low, row, paristech] [multinomial, estimated, binomial, real, classical] [function, error, divergence, logistic, gradient, convex, problem, theorem, method, descent, coordinate, convergence, rate, constrained, estimator, penalized, negative, normalized, lifted, assume, regularization, denoted]
Inference by Learning: Speeding-up Graphical Model Optimization via a Coarse-to-Fine Cascade of Pruning Classifiers
Bruno Conejo, Nikos Komodakis, Sebastien Leprince, Jean Philippe Avouac
Bruno Conejo, Nikos Komodakis, Sebastien Leprince, Jean Philippe Avouac
[solution, energy, map, label, mrf, pruning, grouping, graphical, set, labeling, random, pruned, number, unary, discrete, subset, vertex, graph, progressively, mrfs, binary, ratio, apply, coarser, node, variable, experiment, agreement] [space, optimal, current, general, heuristic, result, future, sequence, strategy, algorithm, study] [image, scale, learning, stereo, feature, based, train, computed, work, training, optical, representation, coarse, applied, trained, accuracy, object, computer, matching] [model, inference, approach, estimate, local, application, markov] [pairwise, minimization, matrix, compute, fast] [time, direct, estimated, represents, dimensionality] [optimization, framework, active, estimation, linear, function, xmap, strong, group, class, optimizing]
Tree-structured Gaussian Process Approximations
Thang D. Bui, Richard E. Turner
Thang D. Bui, Richard E. Turner
[tree, number, experiment, graphical, structure, set, consider, paper, indirect, marginal, region] [conference, algorithm] [training, learning, input, test, datasets, accuracy, accurate, spatial, challenging, structured, based, compared] [gaussian, local, data, inference, model, posterior, chain, process, prior, regression, fitc, covariance, prediction, missing, approximate, smse, imputation, approach, figure, scheme, whilst, likelihood, terrain, full, supplementary, ubk, kuk, vfe, fck, probabilistic, tractably, bayesian, sde, pic, global] [approximation, sparse, stage] [time, audio, signal, range, three, independent] [function, method, computational, machine, large, cost, size, complexity, error, block, quadratic]
Mode Estimation for High Dimensional Discrete Tree Graphical Models
Chao Chen, Han Liu, Dimitris Metaxas, Tianqi Zhao
Chao Chen, Han Liu, Dimitris Metaxas, Tianqi Zhao
[tree, mode, set, labeling, potential, candidate, graphical, partial, number, junction, labelings, ball, label, geodesic, subgraph, graph, discrete, high, connected, vertex, construct, radius, subgraphs, chen, random, node, topographical, interior, disagree, maximum, monotonically, procedure, iij] [algorithm, space, probability, called] [scale, computer, multiple, learning, computed] [local, figure, data, global, distribution, model, density, sample, true, chain, exponential, prediction, estimate, approach, log, computation, international] [compute, denote, smaller, distance, computing, analysis, corresponding, small, point, supported, compare] [three, underlying, time, estimated] [boundary, size, function, problem, method, consistent, estimation, journal, statistical, machine, error, computational, assumption, large, corresponds, dimensional, theorem, assume]
Making Pairwise Binary Graphical Models Attractive
Nicholas Ruozzi, Tony Jebara
Nicholas Ruozzi, Tony Jebara
[partition, graphical, graph, bethe, binary, switching, trbp, cover, edge, belief, reweighted, energy, map, signed, disconnected, marginal, exists, convergent, solution, message, number, intelligence, potential, frustrated, construction, epinions, construct, strength, set, passing, polynomial, attractive, node, variable, collection] [bound, algorithm, upper, lower, general, conference, order, base] [better, computer, original, ieee, work] [model, propagation, temperature, inference, figure, log, local, true, uncertainty, approximate, scheme] [pairwise, approximation, computing, special, exact] [external, free, independent, time, distinct, change, interaction] [function, theorem, convergence, problem, negative, positive, example, converges, convex, machine, statistical]
Augur: Data-Parallel Probabilistic Modeling
Jean-Baptiste Tristan, Daniel Huang, Joseph Tassarotti, Adam C. Pocock, Stephen Green, Guy L. Steele
Jean-Baptiste Tristan, Daniel Huang, Joseph Tassarotti, Adam C. Pocock, Stephen Green, Guy L. Steele
[code, number, runtime, programming, symbolic, set, program] [probability, draw, design, thread] [language, gpu, automatically, test, performance, generating, representation, work, learning, scale] [model, inference, augur, compiler, probabilistic, distribution, generate, gibbs, parallelism, sample, data, val, parallel, regression, sampler, stan, modeling, rewrite, bayesian, conjugacy, cuda, multivariate, compilation, sampling, var, mcmc, predictive, supplementary, scala, generated, larger, allows, figure, likelihood, implementation, yield, sampled, amount, highly] [support] [system, lda, topic, time, collapsed, dirichlet, experimental, single, generates, memory, document, drawn, rule] [large, machine, linear, example, effective, bayes, size, simple, cost, method]
Advances in Learning Bayesian Networks of Bounded Treewidth
Siqi Nie, Denis D. Maua, Cassio P. de Campos, Qiang Ji
Siqi Nie, Denis D. Maua, Cassio P. de Campos, Qiang Ji
[treewidth, milp, graph, set, structure, solution, number, twilp, parent, pit, node, dag, programming, ordering, compatible, topological, integer, dandelion, maximum, collection, clique, maximal, subgraph, constraint, parviainen, community, partial, cplex, provide, hill, nursery, chordal, elimination, directed, korhonen, valid, consider, undirected] [bounded, algorithm, optimal, order, search, good, space] [learning, score, network, yij, work, based, learn] [bayesian, data, sampling, approach, approximate, moral, sample, uncertainty, local, model, inference, breast, exponential, sampled] [formulation, exact, uniformly, refer, small] [time, generation, audio] [method, linear, problem, large, complexity, normalized, size, empirically]
A Statistical Decision-Theoretic Framework for Social Choice
Hossein Azari Soufiani, David C. Parkes, Lirong Xia
Hossein Azari Soufiani, David C. Parkes, Lirong Xia
[set, maximum, number, hard] [decision, social, condorcet, kemeny, voting, alternative, winner, normative, space, ranking, probability, pnp, majority, minimize, focus, optimal, ltop, ranked, satisfy, ariel, minimizes, design, uniform, expected, wmg, selects, order, choose, lirong, dispersion, craig, vincent, objectively, studied, criterion, tyler, main, good, ioannis, procaccia] [composed, work, better, score, truth, pair, ground, natural] [bayesian, model, generated, parameter, parametric, data, likelihood, prior, figure, approach, distribution, mle, sample, david] [denote, top, distance, second, rank] [choice, rule, three] [theorem, statistical, loss, risk, framework, linear, theory, complexity, estimator, function, example, computational, problem, denoted, large, asymptotic]
Transportability from Multiple Environments with Limited Experiments: Completeness Results
Elias Bareinboim, Judea Pearl
Elias Bareinboim, Judea Pearl
[causal, transportability, selection, set, transport, trmz, diagram, exists, complete, consider, collection, limited, observational, interventional, graph, paper, induced, relation, root, aaai, menlo, fact, requirement, edge, bareinboim, nyc, subgraph, computability, characterization, graphical, formal, existence, deciding] [algorithm, conference, return, hedge, result, called, proof, goal, general, appendix, sequence, case] [multiple, learning, computable, pair, computer] [target, data, source, national, model, treatment, university] [knowledge, condition, analysis, heterogeneous] [experimental, represents, state] [domain, relative, structural, formula, problem, theorem, example, machine, statistical, simple]
Randomized Experimental Design for Causal Graph Discovery
Huining Hu, Zhentao Li, Adrian R. Vetta
Huining Hu, Zhentao Li, Adrian R. Vetta
[graph, causal, number, edge, clique, random, adversary, essential, experiment, designer, forcing, arc, discover, maximum, undirected, observational, orient, set, directed, apply, characterization, randomization, oriented, acyclic, skeleton, chordal, vertex, list, cardinality, high, operation, needed, acyclicity, forced, subgraph, remark, collection, maximal, construct, fact, applies, eberhardt, independently, hauser] [bound, case, probability, result, expected, worst, lower, design, selects, simply, commit, upper, access, proof] [entire, based, directional, improve, test, selected] [log, data, model, distribution, markov, supplemental] [required, recover, matrix] [experimental, orientation, typical] [theorem, size, constant, equivalence, randomized, assume, prove, journal]
Poisson Process Jumping between an Unknown Number of Rates: Application to Neural Spike Data
Florian Stimberg, Andreas Ruttor, Manfred Opper
Florian Stimberg, Andreas Ruttor, Manfred Opper
[number, path, discrete, set, continuous, high, switching, random] [probability, chosen, unknown, algorithm, current, order] [based, adding, segment] [data, posterior, process, sampler, model, prior, distribution, figure, inference, mcmc, bayesian, true, markov, likelihood, proposal, sampling, gaussian, allows, sample, estimate, neighboring, black] [small, analysis] [time, state, poisson, jump, spiking, orientation, neural, spike, gamma, chinese, drawn, restaurant, change, distinct, inhomogeneous, triangle, burst, neuronal, stimulus, bursting, event, dirichlet, width, single, join, experimental] [rate, journal, function, assume, large, machine, statistical, simple, active]
Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
David Eigen, Christian Puhrsch, Rob Fergus
David Eigen, Christian Puhrsch, Rob Fergus
[set, map, provide, random] [difference, case] [depth, coarse, network, image, training, output, stereo, input, better, conv, rgb, kitti, ground, layer, nyu, scene, fine, scale, dataset, trained, camera, train, learning, well, predict, applied, monocular, task, alignment, additional, test, work, feature, truth, performance, object, view, deep, pool, entire, convolutional, nyudepth, based, improve, kinect, original, understanding, hidden, correspondence] [global, prediction, log, local, model, rmse, full, target, data, figure, average] [provided, estimating, component, compare] [single, system, raw, qualitative, effectively] [error, method, relative, loss, size, threshold, note, achieves, large]
Causal Strategic Inference in Networked Microfinance Economies
Mohammad T. Irfan, Luis E. Ortiz
Mohammad T. Irfan, Luis E. Ortiz
[causal, program, set, increase, solution, exists, constraint, high, existence, graphical, complete] [equilibrium, mfi, interest, village, mfis, loan, strategic, proof, case, lemma, best, algorithm, study, total, market, appendix, max, economic, repayment, revenue, lower, social, search, maximize, grameen, bolivia, removal, lending, networked, closely, demand, game, bangladesh, evidence] [learning, bank, based, learned, development] [model, amount, inference, data, modeling, university, scheme, log, equation, parameter] [point, special, compute, entry, formulation] [change, response, observed, property, noise, generation, subject] [rate, objective, empirical, function, term, optimization, journal, group, theory, theorem, convex]
Asymmetric LSH (ALSH) for Sublinear Time Maximum Inner Product Search (MIPS)
Anshumali Shrivastava, Ping Li
Anshumali Shrivastava, Ping Li
[lsh, hash, mips, hashing, asymmetric, alsh, srp, construct, item, number, user, locality, random, maximum, interested, recommender, fact, shrivastava, collaborative, structure, paper, set] [search, space, probability, algorithm, max, sublinear, best] [based, proposed, similarity, object, coding, detection, well, transformation] [data, log, approximate, scheme, figure, latent, standard, precision, popular] [vector, product, top, computing, movielens, point, norm, technical, matrix, provably, approximation, distance, compute, corresponding] [time, recall, event, independent, collision] [inner, query, function, neighbor, problem, arg, solving, sensitive, family, framework, theorem, class, theoretical, instance, linear, key, practical]
Hamming Ball Auxiliary Sampling for Factorial Hidden Markov Models
Michalis Titsias, Christopher Yau
Michalis Titsias, Christopher Yau
[hamming, ball, binary, consider, set, energy, number, radius, variable, factor, asymmetric] [algorithm, sequence, current, exploration, uniform, space, probability] [hidden, based, joint, test, training, jointly, applied, location] [sampling, data, model, gibbs, local, posterior, distribution, figure, markov, conditional, step, sample, day, fhmm, auxiliary, inference, disaggregation, mse, factorial, standard, sampler, approach, generated, latent, restricted, trapped, bayesian, density, chain, electricity, xtrue, full, allow, mcmc, scheme] [exact, vector, matrix, distance, corresponding, row, small, column] [time, state, observed, three, single] [block, complexity, example, statistical, locally, updating]
Log-Hilbert-Schmidt metric between positive definite operators on Hilbert spaces
Minh Ha Quang, Marco San Biagio, Vittorio Murino
Minh Ha Quang, Marco San Biagio, Vittorio Murino
[mathematical, structure, apply, induced, strictly, paper, motivated, set] [space, case, version, main, general, bounded, current, difference] [manifold, image, input, dataset, original, computed, learning, based, applied, work, feature, outperforms, computer, novel, identity, task] [covariance, kernel, riemannian, approach, gaussian, log, data, standard, straightforward] [loghs, distance, hilbert, operator, corresponding, matrix, vector, product, stein, texture, compute, dloghs, symmetric, jeffreys, denote, unitized, norm, extended, formulation, algebra, material, ehs, gram, euclidean, generalize, implicitly, condition] [scalar] [positive, metric, inner, rkhs, framework, theorem, bregman, class, computational, assume, linear, machine, including, size, generalization, empirical, regularized]
Learning Optimal Commitment to Overcome Insecurity
Avrim Blum, Nika Haghtalab, Ariel D. Procaccia
Avrim Blum, Nika Haghtalab, Ariel D. Procaccia
[set, number, region, loop, membership, attack, radius, feasible, ball, subset, potential, interior, polynomial, rational, procedure] [strategy, optimal, probability, coverage, algorithm, lemma, attacker, defender, security, game, stackelberg, initial, payoff, pure, attacking, attacked, main, space, resource, best, implementable, utility, proof, result, intersection, letchford, leader, conference, receives, allocation, conservative, highest, tauman, search, total, call, double, denominator, kalai, commit] [representation, learning, description] [target, approach, restricted, step, international, university] [point, vector, compute, requires, determine] [response] [optimization, theorem, linear, convex, randomized, method, margin, query, feasibility, assumption, problem, assume, optimize]
Diverse Randomized Agents Vote to Win
Albert Jiang, Leandro Soriano Marcolino, Ariel D. Procaccia, Tuomas Sandholm, Nisarg Shah, Milind Tambe
Albert Jiang, Leandro Soriano Marcolino, Ariel D. Procaccia, Tuomas Sandholm, Nisarg Shah, Milind Tambe
[number, set, tree, random, quality, making, move, increase] [voting, agent, team, ranking, uniform, winning, alternative, probability, plurality, vote, appendix, condorcet, best, called, fuego, search, marcolino, social, borda, board, neutral, order, proof, main, maximin, consisting, result, ranked, position, mild] [diverse, truth, ground, computer, output, multiple, novel, outperforms, performance, diversity, view, composition] [model, true, distribution, monte, carlo, figure, sample, sampled, sampling] [noisy, denote, compare, rank, copy, pairwise, corresponding] [noise, state, rule, single, choice, presented] [theoretical, biased, denoted, constant, theorem, large, randomized, rate, consistent]
Streaming, Memory Limited Algorithms for Community Detection
Se-Young Yun, marc lelarge, Alexandre Proutiere
Se-Young Yun, marc lelarge, Alexandre Proutiere
[green, asymptotically, graph, classify, red, number, indirect, accurately, partition, proportion, community, grows, requirement, avw, edge, linearly, exists, store, partial, random, construct, connected, node, limited, consider, typically, set, existence, constitutes, awv] [algorithm, case, online, probability, order, previous, observe, general, depend] [accurate, network, detection, output, performance, better, computer, task] [model, data, scaling, perform, step, log] [clustering, matrix, streaming, spectral, column, corresponding, requires, condition, adjacency, randomly, reconstruct, denote, largest, sparse, revealed, required] [memory, observed, cluster, presented] [size, block, problem, theorem, assume, note, stochastic, address, large, statistical, require, arg, stored]
Computing Nash Equilibria in Generalized Interdependent Security Games
Hau Chan, Luis E. Ortiz
Hau Chan, Luis E. Ortiz
[ordering, variable, set, number, exists, consider, graph, mechanism, indirect, reduce, assignment, potential, polynomial, partial] [invest, player, psne, game, clause, play, msne, nash, investment, apartment, kearns, extinguishing, ortiz, probability, case, security, algorithm, determining, equilibrium, investing, interdependent, lemma, decision, study, playing, safety, strategic, extension, michael, agent, best, general, conference, heal, uniform, negated, type, club, karate] [transfer, based, traditional, table] [model, parameter, figure, wrt] [compute, computing, corresponding, denote, material] [direct, time, response, event] [risk, consistent, problem, complexity, generalized, theorem, computational, function, note, suppose, cost, replace, variant]
Approximating Hierarchical MV-sets for Hierarchical Clustering
Assaf Glazer, Omer Weissbrod, Michael Lindenbaum, Shaul Markovitch
Assaf Glazer, Omer Weissbrod, Michael Lindenbaum, Shaul Markovitch
[tree, number, connected, high, set, structure, hierarchy, construct, graph, left, solution, olive, correspond, genetic, oil, evolutionary, recursively, calculated, nne, bikmeans, divisive, existing, program, spanning, chaining, quality, john] [associated, probability, side, algorithm, bandwidth] [level, feature, test, human, train, proposed, learning, dataset, training] [density, data, figure, approximated, kde, approach, estimate, sampled, mass, parallel] [clustering, estimating, synthetic, corresponding, vector, analysis] [cluster, hierarchical, estimated, single, population, european, modal] [method, class, nearest, respect, journal, estimator, function, large, active, neighbor, measure]
Tight Continuous Relaxation of the Balanced k-Cut Problem
Syama Sundar Rangapuram, Pramod Kaushik Mudrakarta, Matthias Hein
Syama Sundar Rangapuram, Pramod Kaushik Mudrakarta, Matthias Hein
[relaxation, continuous, balanced, graph, solution, cut, balancing, bcut, ratio, membership, set, cheeger, partition, asymmetric, rounding, lovasz, monotonic, tight, label, strictly, mtv, feasible, paper, exists, partitioning, corollary, fiji, number, fact, constraint, existing] [best, algorithm, case, optimal, extension, order, submodular, total, minimizing, guarantee, observe, sequence, variation] [proposed, based, better, datasets, work, performance, well] [approach, yield] [clustering, spectral, matrix, symmetric, exact, greedy, nonnegative, simplex, denote, component, transductive, corresponding, minimization, column, row] [directly, subject, choice, wij] [problem, method, function, note, descent, convex, min, size, optimization, theorem, corresponds, normalized, solving]
Learning Multiple Tasks in Parallel with a Shared Annotator
Haim Cohen, Koby Crammer
Haim Cohen, Koby Crammer
[binary, number, label, set, random, corollary] [algorithm, feedback, online, uniform, setting, exploration, round, contextual, bound, total, exploit, focus, exploitation, bandit, good, best, observe, case, receives, version, appears] [task, test, learning, annotator, training, input, mnist, work, performs, evaluated, multiple, multiclass, output, selective, labeled, performance, text, score] [prior, prediction, data, perform, average, model, true] [second, reduction, analysis, denote, multitask] [single, shared, fraction, three, document] [shampo, error, aggressive, problem, xjt, yjt, update, machine, usps, queried, mistake, assume, harder, query, claudio, linear, margin, wjt, perceptron, koby, noting]
Multitask learning meets tensor factorization: task imputation via convex optimization
Kishan Wimalawarne, Masashi Sugiyama, Ryota Tomioka
Kishan Wimalawarne, Masashi Sugiyama, Ryota Tomioka
[mode, set, number, random, factor] [setting, case, result, order, bounded, minimizer, probability] [learning, task, table, performed, training, performance, better, denoising, higher, proposed, selected] [latent, sample, data, log, figure, advantage, true, allows, mse, priori, gaussian] [trace, norm, tensor, rank, multilinear, multitask, scaled, overlapped, matrix, lowest, completion, yipq, heterogeneous, synthetic, mlmtl, dimension, xipq, low, mtl, tomioka, vector, second, recognizes, mpq, homogeneous, maxk, wpq, recognize, explained, worse] [restaurant, three, weight] [convex, min, error, assume, risk, regularization, squared, size, note, problem, relative, dual, dependence, respect, excess]
On Multiplicative Multitask Feature Learning
Xin Wang, Jinbo Bi, Shipeng Yu, Jiangwen Sun
Xin Wang, Jinbo Bi, Shipeng Yu, Jiangwen Sun
[solution, set, penalty] [optimal, lemma, algorithm, general, proof, minimizes, best, lower, study, chosen] [feature, learning, task, joint, computed, performance, applied, proposed, table, based, entire, selected] [data, model, parameter, figure, sharing, regression, probabilistic, explore] [equivalent, norm, component, multitask, synthetic, matrix, formulation, vector, early, sparse, alternating, sparsity, denote, row] [multiplicative, derived, shared, common, share] [problem, mtfl, formula, theorem, function, method, objective, optimization, solve, regularized, regularizers, shrinkage, regularization, loss, family, framework, respect, regularizer, form, theoretical, empirical, convex, subproblem, optimize, solved, solving, relative, empirically, generalized, normal]
Learning convolution filters for inverse covariance estimation of neural network connectivity
George Mohler
George Mohler
[paper, solution, consider, high, number, connected, random] [winning, algorithm, case] [network, convolution, learning, learned, performance, score, training, supervised, learn, based, accurate] [covariance, model, series, equation, data, figure, sample, parameter, approach, processing, highly, estimate, gaussian, prediction] [matrix, sparse, compute, thresholding, fast, determined, corresponding, provided] [time, inverse, auc, signal, neuron, connectivity, neural, imaging, kaggle, competition, glasso, inferring, brain, derivative, calcium, fti, observed, raw, binomial, connectomics, fluorescence, penalization] [estimation, function, method, loss, size, threshold, journal, optimize, optimizing, regularization, optimization, coordinate, machine, generalized, simple, respect, applying, penalized, functional]
Extracting Latent Structure From Multiple Interacting Neural Populations
Joao Semedo, Amin Zandvakili, Adam Kohn, Christian K. Machens, Byron M. Yu
Joao Semedo, Amin Zandvakili, Adam Kohn, Christian K. Machens, Byron M. Yu
[structure, number, directed, consider, set, needed, john, factor, identify, maximum] [study, order] [visual, multiple, well, describe, performed, better, applied, propose, explicitly, proposed] [latent, covariance, model, data, observation, canonical, probabilistic, figure, david] [analysis, reduction] [pcca, dimensionality, time, population, activity, neural, interaction, glara, observed, recorded, brain, interact, state, temporal, correlation, aij, neuron, capture, area, spike, simultaneously, succinct, plotted, account, peak, recording, experimental, latents, neuronal, noise] [note, augmented, function, applying, comparing, group, relative, department]
Analysis of Brain States from Multi-Region LFP Time-Series
Kyle R. Ulrich, David E. Carlson, Wenzhao Lian, Jana S. Borg, Kafui Dzirasa, Lawrence Carin
Kyle R. Ulrich, David E. Carlson, Wenzhao Lian, Jana S. Borg, Kafui Dzirasa, Lawrence Carin
[set, number, region, assignment, consider] [lower, transition, environment, chosen, bound, probability, considered, algorithm, max] [proposed, split, novel, work, multiple, represented, hidden] [model, data, mixture, variational, inference, gaussian, process, kernel, density, distribution, generated, covariance, markov, latent, figure, observation, global, allow, generative, toy, local, bayesian, posterior, prior, predictive] [spectral, tensor, factorization, analysis, vector, component, recovered] [brain, state, cluster, clust, lfp, sleep, animal, time, activity, merge, single, content, dirichlet, arousal, window, drawn, truncation, merges, share, inferred, neural, usage, represent, electrophysiological, shared, lfps, vta] [method, function, iteration]
Spike Frequency Adaptation Implements Anticipative Tracking in Continuous Attractor Neural Networks
Yuanyuan Mi, C. C. Alan Fung, K. Y. Michael Wong, Si Wu
Yuanyuan Mi, C. C. Alan Fung, K. Y. Michael Wong, Si Wu
[continuous, consider, mechanism, wide, localized, strength] [exp, position, feedback, study, current] [network, input, motion, visual, employ, representation, object, proposed] [model, gaussian, larger, figure, track] [leading, condition, support, separation] [neural, tracking, sfa, anticipative, neuronal, travelling, time, moving, ext, speed, wave, external, cann, stimulus, bump, asymmetrical, vext, perfect, interaction, intrinsic, head, brain, vint, state, attractor, range, compensate, amplitude, implement, motor, response, synaptic, canns, mobility, anticipation, effectively, lagging, property, rodent, science, dynamical, induces, system, drive, activity, observed, experimental, orientation, recurrent, command, internal, cognitive] [direction, negative, constant, example, delay, effective, written, key]
Real-Time Decoding of an Integrate and Fire Encoder
Shreya Saxena, Munther Dahleh
Shreya Saxena, Munther Dahleh
[provide, causal, number, consider] [bound, bounded, algorithm, upper, order, case, bandwidth, feedback, current, satisfy, result] [reconstruction, input, applied, output, ieee, well, recursive, accurate] [model, estimate, figure, density, integral] [norm, introduce, knowledge, reconstruct, corresponding, analysis, provided] [signal, decoder, time, decoding, iaf, spike, encoding, neuron, ftk, amplitude, encoder, reconstructed, control, decay, fti, extrinsic, polynomially, integrate, neural, stability, system, innovation, decoded, nonlinear, perfect, hodgkin, prosthetic, membrane, fire, bandlimited, encoded, record] [error, linear, theorem, constant, increasing, example, rate, theory, depends, function, theoretical, numerically, threshold, journal, practical, convergence]
Fairness in Multi-Agent Sequential Decision-Making
Chongjie Zhang, Julie A. Shah
Chongjie Zhang, Julie A. Shah
[solution, programming, number, set, sum, centralized, existing, strictly] [fairness, policy, maximin, optimal, mdps, criterion, algorithm, factored, game, resource, utilitarian, best, agent, action, reward, nash, deterministic, decentralized, dec, mdp, fair, allocation, proposition, division, total, probability, maximizing, decision, payoff, space, equilibrium, conference, maximize, utility, current, minimum, max, pareto, minimizing, opt, termination, exploiting, initial] [performance, joint, work, computed, represented, scale] [approach, approximate, scalable, develop, model, local, international, computation, exponential, sequential] [iterative, computing, compute, small] [state, response, derived, individual] [linear, regularized, function, problem, large, stochastic, min, solving, optimization, assume, simple]
Repeated Contextual Auctions with Strategic Buyers
Kareem Amin, Afshin Rostamizadeh, Umar Syed
Kareem Amin, Afshin Rostamizadeh, Umar Syed
[number, consider, high, fact, set, apply, symposium] [buyer, algorithm, regret, truthful, price, seller, round, lemma, bound, surplus, leap, setting, probability, revenue, online, phase, good, valuation, exploit, repeated, proof, reserve, contextual, market, context, discount, lie, bid, strategic, result, offered, auction, sequence, sublinear, optimal, kernelized, lower, minimize, offer, proposition, advertising, order, observe, case, pricing, accepts, difference, goal, unknown, behaves, future, minimizing] [learning, multiple] [parameter, model, estimate] [analysis, second, vector, introduce] [single, observed, behavior, choice] [function, gradient, note, convex, stochastic, loss, assume, descent, prove, linear, assumption, relative, theorem]
Discovering, Learning and Exploiting Relevance
Cem Tekin, Mihaela Van Der Schaar
Cem Tekin, Mihaela Van Der Schaar
[set, interval, relation, high, number, consider, paper, subset] [reward, action, context, type, regret, relevance, relevant, expected, learner, bandit, algorithm, contextual, bound, drel, exploitation, order, online, case, observe, relt, probability, variation, space, conference, feedback, best, called, exploration, controlled, exploit, tuples, lower, decision, general, depend, sublinear, benchmark, vart, choose] [learning, based, select, selected, work, performance, level, pair, learn, propose] [sample, parameter, international, explore, prior] [vector, pairwise, dimension, denotes, corresponding] [time, estimated, duration, control, independent] [problem, depends, theorem, assumption, linear, dimensional, form, lipschitz, active, denoted, calculate, machine]
On the Computational Efficiency of Training Neural Networks
Roi Livni, Shai Shalev-Shwartz, Ohad Shamir
Roi Livni, Shai Shalev-Shwartz, Ohad Shamir
[polynomial, number, consider, set, corollary, mapping, hard, solution, apply, runtime] [algorithm, case, proof, result, easy, focus, goal, interesting, good, max, associated, previous] [activation, network, learning, depth, training, hidden, relu, geco, input, architecture, layer, output, learned, learn, based, trained, idea, accuracy, train, expressiveness, well] [data, approximate, sample, global, computationally, university, target, observation, step] [approximation, greedy, vector, small, boolean, supported, eigenvalue, dimension] [neural, time, neuron, express] [function, class, complexity, threshold, hardness, theorem, size, sgd, hypothesis, computational, linear, sigmoidal, constant, large, margin, loss, problem, assume, implemented, positive, squared, optimization, machine, convex, note, empirically]
Multilabel Structured Output Learning with Random Spanning Trees of Max-Margin Markov Networks
Mario Marchand, Hongyu Su, Emilie Morvant, Juho Rousu, John S. Shawe-Taylor
Mario Marchand, Hongyu Su, Emilie Morvant, Juho Rousu, John S. Shawe-Taylor
[def, tree, graph, spanning, random, complete, set, rta, consider, number, scoring, wti, yti, edge, collection, multilabels, high, exists, mam, john, microlabel, marginal, map, paper, slack, potential] [lemma, best, probability, highest, algorithm, space, guarantee] [output, feature, unit, learning, joint, score, datasets, combination, training, structured, ensemble, input, mit, based] [inference, sample, approach, approximate, markov, supplementary, average, prediction, expectation, maximizer] [vector, norm, exact, denote, conical, small, material, provably] [weight, simultaneously, individual] [margin, multilabel, predictor, problem, function, loss, achieves, example, machine, optimization, argmax, theorem, large, note, ramp, empirical, dual, mmcrf, normalized, generalization, min]
Metric Learning for Temporal Sequence Alignment
Rémi Lajugie, Damien Garreau, Francis Bach, Sylvain Arlot
Rémi Lajugie, Damien Garreau, Francis Bach, Sylvain Arlot
[hamming, set, graph, rta, program, number, warping, constraint] [best, goal, algorithm, max, good, mahalanobis, setting, context] [alignment, learning, learn, performance, learned, combination, structured, ltb, feature, propose, dataset, evaluation, task, symmetrized, similarity, better, pair, groundtruth, cast, work, ieee, training, proposed, output] [prediction, series, perform, standard, data, figure, parameter] [matrix, leading] [area, time, audio, lta, music, dynamic, signal, decoding, temporal, real, single, depicted, length, aligning, mfccs, speech] [loss, metric, problem, linear, note, framework, optimization, large, measure, margin, handcrafted, mfcc, violated, simple, solve, convex, conducted, corresponds, journal, function]
Proximal Quasi-Newton for Computationally Intensive L1-regularized M-estimators
Kai Zhong, En-Hsu Yen, Inderjit S. Dhillon, Pradeep K. Ravikumar
Kai Zhong, En-Hsu Yen, Inderjit S. Dhillon, Pradeep K. Ravikumar
[set, number, labeling, label, partial, solution, random] [algorithm, sequence, strategy, oracle, current, appendix, search, optimal, space, criterion] [feature, crf, learning, evaluation, dataset] [data, inference, model, conditional, computation, figure, restricted, computationally] [sparsity, requires, second, fast, vector, hessian, sparse, approximation, condition, corresponding] [time, hierarchical, raw, memory] [proximal, gradient, convergence, method, update, working, problem, inner, coordinate, descent, strong, shrinking, function, superlinear, active, sgd, expensive, key, instance, constant, complexity, objective, asymptotic, relative, aggressive, convex, bcd, intensive, updated, convexity, large, nullspace, variant, size, iteration, iterate, carefully, optimization, theorem, loss]
Discriminative Metric Learning by Neighborhood Gerrymandering
Shubhendu Trivedi, David Mcallester, Greg Shakhnarovich
Shubhendu Trivedi, David Mcallester, Greg Shakhnarovich
[set, number, label, binary] [algorithm, max, mahalanobis, goal, optimal, majority, highest] [learning, correct, structured, training, work, score, based, accuracy, similarity, surrogate, learned, output, neighborhood, labeled, learn, feature, combination, circle] [inference, latent, target, approach, prediction, data, allow, standard, model, allows] [distance, euclidean, point, formulation, compact, exact, scaled, matrix, frobenius] [choice, direct] [metric, loss, class, knn, nearest, method, objective, gradient, optimization, lmnn, note, stochastic, neighbor, augmented, large, including, form, incorrect, isolet, mlr, margin, descent, regularizer, optimize, update, involve, intuition, nca, targeted, corresponds, usps, gerrymandering, itml, problem]
Fundamental Limits of Online and Distributed Algorithms for Statistical Learning and Estimation
Ohad Shamir
Ohad Shamir
[consider, partial, number, apply, set, provide] [online, lower, algorithm, bound, learner, feedback, bandit, type, detect, general, optimal, context, regime, setting, bounded, regret, case, implies, depend, upper, access, proof, goal] [learning, performance, work, based] [data, sample, covariance, standard, distribution, larger, bias] [sparse, pca, vector, detecting, requires, special, small, smaller, provably] [memory, independent, mutual, interact, state, include, distinct] [distributed, stochastic, coordinate, statistical, problem, estimation, size, protocol, complexity, optimization, communication, machine, theorem, loss, biased, example, linear, simple, empirical, prj, convex, prove]
Information-based learning by agents in unbounded state spaces
Shariq A. Mobin, James A. Arnemann, Fritz Sommer
Shariq A. Mobin, James A. Arnemann, Fritz Sommer
[number, set, procedure, consider, discrete] [agent, reward, unbounded, action, maze, exploration, bounded, probability, transition, environment, policy, crp, optimal, reinforcement, unknown, strategy, gain, current, competitor, space, planning, best, padding, absorbing, dklp, observable, cmc, order, adjusted, cumulative, literature, previous, explorative, undiscovered, assessing, algorithm, follow] [learning, proposed, learn, based, predicted, discovered, table] [model, missing, figure, sampling, process, bayesian, prior, parameter, chain, markov, target, explore, david, true] [smaller, calculating] [state, time, internal, described, neural, chinese, restaurant, dirichlet, lta, describes] [iteration, problem, bayes, size, objective, address, empirical]
Learning on graphs using Orthonormal Representation is Statistically Consistent
Rakesh Shivanna, Chiranjib Bhattacharyya
Rakesh Shivanna, Chiranjib Bhattacharyya
[graph, ersn, transduction, random, set, high, number, notion, subgraph, exists, edge, undirected, corollary, binary, ysn, relating, degree, law] [algorithm, probability, bound, associated, combining, study, optimal, result, setting, uniform] [learning, multiple, representation, performance, novel, unit, propose, embedding] [kernel, processing, sample, data, kmm, style] [orthonormal, denote, analysis, transductive, laplacian, corresponding, formulation] [fraction, dependent, neural, inverse, choice] [complexity, class, labelled, rademacher, large, consistency, function, theorem, derive, generalization, problem, error, machine, consistent, note, loss, ramp, labelling, convergence, margin, statistical, structural, measure, min, constant, rate, family, working, hinge, solving]
Optimal prior-dependent neural population codes under shared input noise
Agnieszka Grabska-Barwinska, Jonathan W. Pillow
Agnieszka Grabska-Barwinska, Jonathan W. Pillow
[constraint, energy, number, consider, induced, grows] [optimal, bound, total, space, lower, optimality, order, gray, literature, allowed, counting] [input, coding, fisher, entire, work, based] [prior, mse, variance, gaussian, bayesian, log, posterior, figure, likelihood, average, model, tractable, processing, true, allows, amount, inference] [exact, small, second] [tuning, population, width, neural, noise, stimulus, spike, poisson, count, time, mutual, curve, amplitude, tiling, narrow, preferred, idealized, examine, derived, window, response, orientation, independent, sensory, correlation, oct, shared, stdev, neuronal, range, spacing] [note, depends, function, squared, loss, error, rate, form, large, constant]
Distributed Variational Inference in Sparse Gaussian Process Regression and Latent Variable Models
Yarin Gal, Mark van der Wilk, Carl Rasmussen
Yarin Gal, Mark van der Wilk, Carl Rasmussen
[variable, set, node, partial, number, marginal, maximum] [algorithm, case, lower, bound, central, utility, optimal] [dataset, training, learning, proposed, scale, performance, well, mnist, trained, datasets, better, improves, additional, test, based, performed] [inference, data, regression, latent, gaussian, inducing, log, model, process, variational, distribution, big, parallel, titsias, perform, supplementary, svi, global, figure, scaling, load, allows, gps, optimisation, computation, modelling, kmi, gplvm, full, standard, likelihood, derivation, local, posterior, demonstrate, flight, scg, prior] [sparse, running, material, scaled, ridge, approximation] [time, observed] [computational, distributed, large, function, machine, increasing, linear, showing, stochastic]
Improved Distributed Principal Component Analysis
Yingyu Liang, Maria-Florina F. Balcan, Vandana Kanchanapally, David Woodruff
Yingyu Liang, Maria-Florina F. Balcan, Vandana Kanchanapally, David Woodruff
[solution, number, set, symposium, factor, coresets, quality, acm, node] [algorithm, probability, choose, lemma, proof, annual, case] [embedding, mnist, original, work, embeddings, computed, computer] [data, global, local, approximate, perform, figure, log] [pca, subspace, matrix, svd, principal, dispca, clustering, projection, dimension, approximation, singular, low, projected, rank, component, fast, compute, analysis, denote, orthonormal, top, point, success, small, second, reduction, spanned, computing, row, speedup, improved, basis, union, running, embedded, coordinator] [time, property, newsgroups, directly] [distributed, communication, randomized, cost, error, theorem, linear, large, note, server, computational, constant, dimensional]
Bregman Alternating Direction Method of Multipliers
Huahua Wang, Arindam Banerjee
Huahua Wang, Arindam Banerjee
[penalty, gurobi, set, consider, factor, variable, program, bethe, composite, runtime, augmentation, structure, graph] [transportation, case, algorithm, commercial, optimality, choosing, version, choose] [unit, additional, based, propose] [global, mass, step, figure, parallel, university, highly] [alternating, faster, residual, special, projection, siam] [memory, three, optimized] [badmm, bregman, admm, divergence, quadratic, term, convergence, generalized, bzt, method, objective, problem, function, update, solving, linear, direction, convex, gradient, descent, dual, proximal, solved, solve, mirror, complexity, size, linearizing, mda, lagrangian, iteration, converges, journal, kkt, replace, primal, equality, generalizes, min, note, logistic, theorem, large, machine, exponentiated, framework]
On Model Parallelization and Scheduling Strategies for Distributed Machine Learning
Seunghak Lee, Jin Kyu Kim, Xun Zheng, Qirong Ho, Garth A. Gibson, Eric P. Xing
Seunghak Lee, Jin Kyu Kim, Xun Zheng, Qirong Ho, Garth A. Gibson, Eric P. Xing
[worker, partial, set, number, programming, graphlab, subset, store, list] [algorithm, return, order, pull, stale, strategy, concurrent, access] [learning, multiple, based] [model, parallel, data, parameter, schedule, big, figure, scheduling, push, parallelization, larger, global, latent, yahoolda, master, sampling, counter, allow, execute, implementation, modeling, distribution, gibbs, partitioned, full, allows, inference] [matrix, factorization, compute, small, row] [lda, topic, system, speed, dynamic, memory, cluster, observed, time] [lasso, convergence, distributed, update, updated, machine, large, coordinate, objective, error, descent, size]
New Rules for Domain Independent Lifted MAP Inference
Happy Mittal, Prasoon Goyal, Vibhav G. Gogate, Parag Singla
Happy Mittal, Prasoon Goyal, Vibhav G. Gogate, Parag Singla
[map, assignment, variable, set, existing, solution, consider, procedure, corollary, remaining, subset, reduce, notion, structure, belief, number] [algorithm, extreme, exploiting, optimal, lemma, version, order, return, counting, base] [ground, applied, network, original, work] [inference, markov, probabilistic, approach, figure] [second, denote] [single, rule, reduced, time, weight, independent] [theory, mln, domain, occurrence, equivalence, lifted, size, xmap, class, formula, lifting, theorem, problem, nslgrb, identical, predicate, solgrb, solver, logic, decomposer, grb, constant, mlns, tautology, sarkhel, cost, weighted, fte, lift, assume, wmj, complexity, argument, normal, upto, transitivity]
On Integrated Clustering and Outlier Detection
Lionel Ott, Linsey Pang, Fabio T. Ramos, Sanjay Chawla
Lionel Ott, Linsey Pang, Fabio T. Ramos, Sanjay Chawla
[number, relaxation, set, quality, solution, integer, factor, consider, sum, programming, constraint, runtime, ratio, subset, obtains] [algorithm, optimal, selecting, extension, initial] [selected, proposed, detection, impact, location, digit, indicates, evaluation, computed, identity, mnist, jaccard] [data, figure, university, processing, step, larger, dij, model, true] [clustering, outlier, apoc, xij, point, matrix, robust, approximation, facility, synthetic, exactly, column, flo, distance, lof, requires, row, small, compare, principal, determinant, totally, analysis, sydney, equivalent, faster] [time, cluster] [problem, lagrangian, method, cost, linear, subgradient, large, iteration, block, size, theorem, negative, homogeneity, objective, function, convergence]
Constrained convex minimization via model-based excessive gap
Quoc Tran-Dinh, Volkan Cevher
Quoc Tran-Dinh, Volkan Cevher
[set, solution, corollary, construct] [algorithm, gap, optimal, choose, max, bound, order, adaptive, lemma] [accuracy, performance, based] [parameter, data, figure, step, generated] [alternating, point, residual, distance, minimization, siam, numerical, technique, condition, compute, vector, special] [classification, time] [convex, function, convergence, dual, proximal, primal, augmented, theoretical, excessive, feasibility, lagrangian, objective, libsvm, basic, sxc, method, linear, constrained, bregman, direction, update, rate, gradient, min, ame, computational, problem, strong, closed, algorithmic, lipschitz, axk, smooth, empirical, smoothing, solving, smoothed, optimization, practical, framework, inequality]
Learning to Search in Branch and Bound Algorithms
He He, Hal Daume III, Jason M. Eisner
He He, Hal Daume III, Jason M. Eisner
[node, solution, number, pruning, tree, selection, queue, set, scip, root, imitation, programming, integer, feasible, gurobi, pruned, expand, mixed, runtime, milp, fathomed, apply, corlat, subset, mik] [optimal, policy, search, bound, algorithm, oracle, upper, lower, branching, current, order, combinatorial, gap, good, prune, strategy, total, space, adaptive, best, ranked, searching, expected] [learning, training, better, test, learn, based, computed, hybrid, work, dataset, depth, learned, learns, applied, structured, datasets, higher, supervised] [figure, global, approach, prediction, process] [compare] [time, state, dynamic, weight] [problem, solving, linear, method, solver, solved, objective, optimization, rate, class, solve]
Convex Deep Learning via Normalized Kernels
Özlem Aslan, Xinhua Zhang, Dale Schuurmans
Özlem Aslan, Xinhua Zhang, Dale Schuurmans
[number, relaxation, set, consider, solution, structure, provide] [optimal, algorithm, appears, versus, adaptive, case, design] [training, learning, deep, output, jointly, postulate, layer, input, hidden, feature, test, architecture, propensity, transfer, work, deeper, additional] [latent, kernel, model, local, step, figure, data, approach, conditional, allows, modeling] [matrix, formulation, denote, vector, second, improved, totally, corrective, alternating, boolean] [nonlinear, neural, weight, three, single, response, capture, real, expressed] [convex, loss, objective, problem, normalized, linear, min, optimization, error, regularization, key, method, machine, domain, clearly, example, block, large, empirical, margin]
Convex Optimization Procedure for Clustering: Theoretical Revisit
Changbo Zhu, Huan Xu, Chenlei Leng, Shuicheng Yan
Changbo Zhu, Huan Xu, Chenlei Leng, Shuicheng Yan
[membership, solution, set, ratio, procedure, provide, paper, construct, constructed, sum, linearly] [optimal, main, space, proof, conference, case, order, chosen, bound, lemma, result, theoretic, notice, implies, algorithm, difference] [performance, feature, based, learning, proposed] [data, figure, international, sample, university, kernel, larger] [son, matrix, clustering, dimension, denote, cube, distance, row, analysis, column, determine, norm, concatenating, ith, singapore, provided, equivalent, operator, huan, repeat, rigorous, concatenate] [cluster, centered, drawn, hierarchical] [problem, theoretical, convex, empirical, theorem, optimization, size, regularization, term, min, arg, denoted, depends, solve, machine, function, department, linear, respect, form]
Multivariate f-divergence Estimation With Confidence
Kevin Moon, Alfred Hero
Kevin Moon, Alfred Hero
[random, entropy, number, uncorrelated, asymptotically, interval, apply, set] [bound, lemma, central, probability, lower, bounded, proof, conference] [ensemble, ieee, based, unit, work, learning, testing] [density, distribution, limit, multivariate, mse, estimate, inference, data, nonparametric, standard, covariance, bias, var, international, process, sample, parameter, concentration, approach] [support, required] [estimated, neural] [estimator, divergence, estimation, error, theorem, convergence, rate, weighted, asymptotic, hero, machine, bayes, normal, class, assume, cov, optimization, convex, moon, normality, problem, normalized, functional, theory, interchangeable, qda, comparing, optimally, form, applying, consistent, annals, endpoint, functionals, hypothesis]
Parallel Double Greedy Submodular Maximization
Xinghao Pan, Stefanie Jegelka, Joseph E. Gonzalez, Joseph K. Bradley, Michael I. Jordan
Xinghao Pan, Stefanie Jegelka, Joseph E. Gonzalez, Joseph K. Bradley, Michael I. Jordan
[set, graph, maximization, ordering, program, cover, runtime, number, scalability, easily, subset, reduce, cut] [algorithm, max, result, submodular, double, serial, transaction, stale, version, coordination, concurrency, nif, draw, failed, fail, bound, optimal, guarantee, return, zigzag, wait, decrease, commit, monotone, processed, operating, superset, serializable, alse, closely, parallelized, replaced, maximizing, overly, concurrent] [learning, achieved, propose] [parallel, process, true, approach, processing, reduces, lead] [greedy, element, approximation, exact, extended, speedup, multilinear, analysis, implicitly] [control, free, state, time] [min, objective, logical, machine, distributed, cost, server, weak, guaranteed]
From MAP to Marginals: Variational Inference in Bayesian Submodular Models
Josip Djolonga, Andreas Krause
Josip Djolonga, Andreas Krause
[consider, marginals, set, modular, map, random, ield, graph, partition, polymatroid, gmax, tight, entropy, marginal, provide, number, quality, supergradient, experiment, energy, hard] [submodular, lower, general, upper, bound, gap, probability, extension, lemma, case, called, order] [learning, computer, natural, accurate, ieee, accuracy] [log, inference, curvature, variational, bayesian, distribution, scheme, approximate, factorized, probabilistic, approach, model, average, sampled, processing, water, prior, data] [minimization, point, approximation, computing, exact, requires, pairwise, greedy] [capture, single] [function, problem, optimization, note, machine, convex, optimizing, error, cost, pick, optimize, journal, large]
Stochastic Network Design in Bidirected Trees
xiaojian wu, Daniel R. Sheldon, Shlomo Zilberstein
xiaojian wu, Daniel R. Sheldon, Shlomo Zilberstein
[runtime, increase, tree, vertex, number, edge, rounded, list, programming, path, set, directed, maximization, rounding, quality, subtree, hard, intelligence, recursively, remove, arbitrary, apply] [rdp, algorithm, policy, river, expected, reward, bidirected, tuple, passabilities, tuples, design, barrier, habitat, conservation, action, total, budget, reachable, probability, case, bound, upper, subpolicy, landscape, optimal, planning, conference, associated, removal, downstream, daniel, lemma, repair, start, upstream, goal, spread, discretized, bounded, passage] [network, better, original] [model, figure, approach, amount] [greedy, condition, approximation, small] [dynamic, time, connectivity, three] [stochastic, problem, error, theoretical, large, assumption, theorem, cost]
Tighten after Relax: Minimax-Optimal Sparse PCA in Polynomial Time
Zhaoran Wang, Huanran Lu, Han Liu
Zhaoran Wang, Huanran Lu, Han Liu
[procedure, high, relaxation, number, constraint, consider, existing] [algorithm, optimal, main, suboptimal, sequence, bound, minimax, illustrates, upper, total] [learning, proposed, level, propose, novel] [log, covariance, sample, parameter, step, figure] [sparse, principal, subspace, matrix, analysis, orthogonal, pca, sparsity, leading, initialization, component, top, basin, attains, nonconvex, pursuit, denotes, iterative, spiked, tmin, orthonormal, uinit, dimension, nonzero, stage, init, projection, thin, estimating, eigenvectors, spanned, remind, power, denote, row, norm, smaller] [attraction, population] [statistical, rate, optimization, iteration, error, convex, estimator, method, computational, theorem, assumption, annals, journal, problem, size, note, convergence, named, theoretical, solving, term, estimation, large]
Consistency of weighted majority votes
Daniel Berend, Aryeh Kontorovitch
Daniel Berend, Aryeh Kontorovitch
[random, high, independently, consider, making, potential, set, variable] [decision, expert, probability, optimal, exp, majority, opt, committee, bound, agent, lemma, whc, proof, adaptive, competence, frequentist, case, unknown, open, nonadaptive, main, vote, deviation, regime, setting, voting, result, competent, observe, aggregation] [correct, work, based, challenging, learning] [approach, log, bayesian, estimate, highly, concentration, model, sample, standard, mass] [analyze, small, ith, condition, knowledge, sharp, estimating, analysis] [rule, choice, estimated, event, independent] [theorem, empirical, weighted, error, inequality, consistency, min, simple, large, theory, empirically, dependence, function, problem, sign, statistical, term, rate, bernstein, department]
Structure learning of antiferromagnetic Ising models
Guy Bresler, David Gamarnik, Devavrat Shah
Guy Bresler, David Gamarnik, Devavrat Shah
[graph, set, graphical, structure, ising, repelling, antiferromagnetic, degree, subset, edge, consider, random, number, maximum, discrete, high, paper, strength, simplehc, cult, strongrepelling, equal, bento, ferromagnetic, node, undirected, solution, potts, partition] [algorithm, probability, general, lemma, lower, bound, access, bounded, search, result, cient, proof, order, focus] [learning, learn, work, based, question] [model, distribution, log, markov, sample, unbiased, average, computation, estimate, true, likelihood] [denote, pairwise, introduced, requires, second] [correlation, decay, independent, property, interaction, underlying] [statistical, complexity, computational, theorem, require, class, problem, simple, size, journal, family, exponent]
Near-Optimal-Sample Estimators for Spherical Gaussian Mixtures
Ananda Theertha Suresh, Alon Orlitsky, Jayadev Acharya, Ashkan Jafarpour
Ananda Theertha Suresh, Alon Orlitsky, Jayadev Acharya, Ashkan Jafarpour
[consider, set, number, candidate, discrete, high, polynomial, def, collection, factor, reduce, construct, maximum] [algorithm, lemma, probability, bound, bounded, studied, search, space] [learning, recursive, coarse, natural, based] [mixture, gaussian, sample, distribution, log, step, gaussians, spherical, estimate, variance, earn, cheffe, scheffe, odified, data, quantizing, approach, covariance, exhaustive, average] [distance, span, spectral, clustering, close, component, matrix, requires, separation, estimating, analysis, small, largest, running, eigenvectors, credit, top] [time, cluster, underlying, correlation] [complexity, pac, estimator, simple, dimensional, constant, estimation, error, note, convergence, problem, example, derive, class]
Parallel Direction Method of Multipliers
Huahua Wang, Arindam Banerjee, Zhi-Quan Luo
Huahua Wang, Arindam Banerjee, Zhi-Quan Luo
[number, overlapping, composite, set, consider, variable] [setting, choose, algorithm, optimality] [better, multiple] [step, parallel, backward, converge, separable, generated, global] [matrix, randomly, minimization, faster, alternating, component, principal, robust, analysis, sparsity, sparse, residual, special, exactly, small] [ascent, time, direct, three] [pdmm, block, dual, admm, method, coordinate, randomized, update, sadmm, xjt, convergence, direction, convex, rbsumm, solve, gsadmm, problem, group, iteration, primal, xtt, linear, solved, complexity, bregman, proximal, solving, gradient, converges, objective, descent, kkt, pjadmm, considers, optimization, constant, bsumm, large, divergence, size, min, note, rmi, theorem, assume, ajt, splitting]
Constant Nullspace Strong Convexity and Fast Convergence of Proximal Methods under High-Dimensional Settings
En-Hsu Yen, Cho-Jui Hsieh, Pradeep K. Ravikumar, Inderjit S. Dhillon
En-Hsu Yen, Cho-Jui Hsieh, Pradeep K. Ravikumar, Inderjit S. Dhillon
[set, solution, number, typically, composite, consider, feasible] [lemma, optimal, bound, satisfy, algorithm, bounded, sublinear, proof] [long] [step, global, restricted, local, standard, data] [matrix, norm, hessian, condition, distance, second, point, denote, introduce, analysis, sparse, fast, orthogonal, vector, iterative, leverage] [inverse] [proximal, convergence, method, convex, strong, gradient, convexity, linear, statistical, newton, quadratic, constant, function, cnsc, loss, form, descent, nullspace, proxh, decomposable, min, logistic, optimization, theorem, hold, arg, update, iff, note, objective, satisfying, estimation, journal, large, prove, require, suppose, coordinate, problem, positive, machine, variant, proxht, size, iteration]
Quantized Estimation of Gaussian Sequence Models in Euclidean Balls
Yuancheng Zhu, John Lafferty
Yuancheng Zhu, John Lafferty
[random, number, set, high, ball, code, consider, remark, john, procedure] [minimax, lower, bound, result, sequence, case, main, probability, lemma, space, expected, depend, order, limiting, future, budget] [coding, work, codebook, based] [source, nonparametric, figure, data, estimate, log, model, parameter, distribution, computation, computationally, prior, sample, standard, scheme, average, step, generated, naturally] [vector, euclidean, establish, lim, compute] [decoding, encoder, encoding, encode, classical, denoting, decoder] [risk, estimation, quantized, distortion, rate, function, inf, problem, theorem, communication, statistical, theory, normal, estimator, method, suppose, error, loss, lossy, excess, asymptotic, squared, distributed, computational, tradeoff, complexity, term, measure, journal, convergence]
On the Convergence Rate of Decomposable Submodular Function Minimization
Robert Nishihara, Stefanie Jegelka, Michael I. Jordan
Robert Nishihara, Stefanie Jegelka, Michael I. Jordan
[discrete, linearly, corollary, number, graph, set, arbitrary, solution, sum, cut, remark] [submodular, bound, lemma, algorithm, proposition, optimal, friedrichs, appendix, sfm, minimizing, lower, general, bauschke, upper, base, arm, case, relevant, deutsch, combinatorial, extension, implies, uniform] [bounding, work, computer, geometry, based] [equation, approach, processing, log, straightforward] [projection, analysis, alternating, matrix, vector, siam, angle, minimization, singular, approximation, largest, condition, spectral, point, eigenvalue] [relate, neural, time] [rate, convergence, problem, convex, function, linear, converges, theorem, journal, optimization, min, decomposable, solving, simple, method, fmax, machine, dual, objective]
The Bayesian Case Model: A Generative Approach for Case-Based Reasoning and Prototype Classification
Been Kim, Cynthia Rudin, Julie A. Shah
Been Kim, Cynthia Rudin, Julie A. Shah
[set, number, assigned, structure, experiment] [decision, case, order, probability] [feature, accuracy, digit, human, dataset, reasoning, unsupervised, learning, output, learn, task, learned, compared, representation, pixel, better, ground, understanding, truth, powerful] [model, mixture, figure, standard, bayesian, observation, sampling, prediction, data, distribution, generate, generative, latent, gibbs, full, prior, approach, generated, inference, hyperparameters] [subspace, xij, vector, top, randomly] [bcm, cluster, prototype, lda, handwritten, interpretability, topic, produced, dirichlet, recipe, zij, cbr, newsgroups, indicator, statistically, interpretable, subject, underlying, psj, horror, presented, gpsj, explain] [example, machine, measure, note]
Zeta Hull Pursuits: Learning Nonconvex Data Hulls
Yuanjun Xiong, Wei Liu, Deli Zhao, Xiaoou Tang
Yuanjun Xiong, Wei Liu, Deli Zhao, Xiaoou Tang
[graph, set, number, subset, selection] [algorithm] [learning, training, dataset, representation, face, original, selected, table, select, image, extracted, input, recognition, coding, manifold, learn, text, refers, applied, object] [data, model, computation, local, sampling, parameter, volume, kernel, approach, perform, sampled, approximate, global] [matrix, zeta, hull, point, anchor, column, sparse, nonconvex, lsc, simplex, extremeness, vector, small, approximation, diagonal, greedy, xei, zhp, row, spectral, adjacency, singular, cur, decomposition, svd, randomly, computing] [time] [complexity, structural, convex, function, measure, theorem, method, machine, form, large, nearest, problem]
SAGA: A Fast Incremental Gradient Method With Support for Non-Strongly Convex Composite Objectives
Aaron Defazio, Francis Bach, Simon Lacoste-Julien
Aaron Defazio, Francis Bach, Simon Lacoste-Julien
[store, composite, number, consider, apply, provide] [algorithm, case, lemma, proof, version, optimal, order] [table, work, additional, explicitly, separate] [step, variance, figure, parameter, conjugate, unbiased, standard, average, approach, supplementary, tested] [incremental, equivalent, reduction, operator, compute, fast, vector, small, technical, support, row, smaller] [described] [saga, method, gradient, convex, update, sag, proximal, sdca, svrg, primal, convergence, dual, size, stochastic, note, convexity, strong, finito, rate, theory, machine, coordinate, quadratic, function, form, constant, iteration, regulariser, updated, theoretical, linear, require, quantity, tong, descent, storage, stored, practical, aaron, empirical, pick]
Dimensionality Reduction with Subspace Structure Preservation
Devansh Arpit, Ifeoma Nwogu, Venu Govindaraju
Devansh Arpit, Ifeoma Nwogu, Venu Govindaraju
[structure, number, disjoint, arbitrary, subset, random, expression] [algorithm, lie, order, conference, case, proof, lemma, notice] [dataset, pair, table, ieee, computer, face, entire, segmentation, accuracy, proposed, labeled, image, datasets, pie, feature, multiple, cmu, propose, idea] [data, approach, independence, figure, perform, sampled, international, equation, generate, local] [vector, projection, subspace, matrix, principal, reduction, projected, ith, analysis, yale, separated, angle, extended, preservation, basis, union, span, preserving, sii, pca, dim, compute, synthetic, support, randomly, session, corresponding] [independent, dimensionality, real, lda, three] [class, theorem, form, dimensional, linear, margin, gradient, method, assumption, acc, plane, empirical]
Distributed Parameter Estimation in Probabilistic Graphical Models
Yariv D. Mizrahi, Misha Denil, Nando de Freitas
Yariv D. Mizrahi, Misha Denil, Nando de Freitas
[lap, composite, clique, marginal, path, mizrahi, liu, normalised, graph, potential, consensus, ihler, connected, xaq, mrf, random, maximum, induced, graphical, edge, node, undirected, apply, summing, mrfs, set, paper, factor, exists, remark, agreement, structure] [general, algorithm, chosen, proposition, case, result, setting, requiring] [joint, learning, boltzmann, work] [likelihood, parameter, conditional, equation, distribution, estimate, local, gibbs, global, markov, auxiliary, data, probabilistic, figure, gaussian] [condition, vector, corresponding, analysis, provided, denotes] [] [strong, distributed, estimation, respect, consistent, estimator, theorem, machine, statistical, asymptotic, convergence, normalized, class, linear, framework, domain, assumption, argument, suppose, form, theoretical]
Blossom Tree Graphical Models
Zhe Liu, John Lafferty
Zhe Liu, John Lafferty
[blossom, tree, graphical, forest, graph, node, nonparanormal, partial, set, bivariate, number, negentropy, pedicel, undirected, edge, maximum, colored, marginal, high, john, partition, random, differential, arbitrary, structure, entropy, apply, spanning, correspond, construct] [algorithm, selects, proposition, controlling, optimal, order, choose] [based, selected, joint, build] [density, gaussian, nonparametric, estimate, data, true, model, step, kernel, covariance, modeling, figure, multivariate, sample, allows, independence] [matrix, denotes, factorization] [estimated, correlation, glasso, property, modeled, single, weight, simulated] [estimation, lasso, family, estimator, method, form, equality, suppose, dimensional, machine, statistical, assumption, note]
Projecting Markov Random Field Parameters for Fast Mixing
Xianghang Liu, Justin Domke
Xianghang Liu, Justin Domke
[marginal, set, random, number, strength, marginals, ising, mrf, edge, mixed, projecting, limited, consider, graph, solution, trw, belief, paper, induced, attractive] [piecewise, algorithm, bound, general] [original, accuracy, reversed] [gibbs, sampling, mixing, distribution, markov, inference, dependency, variational, approximate, figure, chain, propagation, stationary, parameter, lbp, tractable, rij, step, dij, looseness, projc, converge, rapid, true, average] [projection, norm, matrix, euclidean, fast, spectral, univariate, projected, project, distance, running, compare, operator, vector, pairwise] [time, interaction, zij] [gradient, divergence, error, descent, family, min, size, simple, theorem, optimization, dual, method, solved, arg, inf, solve, function, problem]
Consistency of Spectral Partitioning of Uniform Hypergraphs under Planted Partition Model
Debarghya Ghoshdastidar, Ambedkar Dukkipati
Debarghya Ghoshdastidar, Ambedkar Dukkipati
[number, partitioning, hypergraph, partition, random, set, graph, exists, provide, connected, consider, high] [algorithm, result, uniform, bound, lemma, order, probability, conference, satisfy] [computer, matching, vision, ieee, correct, performance, work, motion, based, learning, higher, pattern, segmentation] [model, approach, log, figure, standard, sampled, true] [spectral, matrix, clustering, tensor, analysis, misclustered, planted, hyperedge, condition, orthonormal, eigenvectors, subspace, hypergraphs, detecting, leading, symmetric, corresponding, decomposition, singular, point, component, product, diagonal, surely, blockmodel, correctly] [cluster] [problem, theorem, consistency, size, theoretical, large, geometric, stochastic, method, machine, note]
Low Rank Approximation Lower Bounds in Row-Update Streams
David Woodruff
David Woodruff
[making, random, fact, corollary, message, consider, typically, set] [algorithm, bound, lower, probability, lemma, space, main, deterministic, goal, write, combining, observe, call] [work, output, unit, word] [data, model, full, amount] [matrix, arbitrarily, small, streaming, bob, alice, approximation, row, rank, low, singular, projection, vector, clarkson, woodruff, rur, alg, second, ghashami, compute, norm, supported, liberty, pythagorean, incremental, top, improved, union, entry, phillips, sends, condition, technical, frobenius] [event, single, memory, conditioned, triangle, direct, distinct] [constant, theorem, problem, communication, showing, inequality, simple, eik, large, note, suppose, form, string]
Sparse PCA via Covariance Thresholding
Yash Deshpande, Andrea Montanari
Yash Deshpande, Andrea Montanari
[high, random, number, set, consider, provide, wainwright] [algorithm, probability, order, appears, phase, result, bound, regime, focus, proof, minimax] [work, table] [covariance, estimate, log, sample, model, figure, kernel, standard, approach, gaussian, data] [thresholding, sparse, support, diagonal, principal, matrix, component, analysis, pca, rank, johnstone, largest, denote, recovery, spectrum, sparsity, compute, eigenvector, second, technical, estimating, succeeds, recovered, knowledge, spiked, amini, norm, spectral, vpc, success] [noise, signal, fraction, spike] [theorem, constant, assume, annals, assumption, estimation, empirical, consistent, computational, problem, large, method, error, threshold, practical, size, theory, estimator, function, asymptotic, note]
A Synaptical Story of Persistent Activity with Graded Lifetime in a Neural System
Yuanyuan Mi, Luozheng Li, Dahui Wang, Si Wu
Yuanyuan Mi, Luozheng Li, Dahui Wang, Si Wu
[mechanism, consider, strength, tsp] [appendix, neutral, order, probability, depending] [network, input, china, long, view, proposed, conventional, wang, natural] [figure, amount, simulation, process] [condition, matrix, point, smaller, close] [persistent, activity, neural, state, lifetime, neuronal, time, stf, stimulus, attractor, system, external, stable, std, synaptic, stp, neuron, memory, silent, prefrontal, three, holding, decay, cortex, membrane, slowly, property, cortical, spike, neuroscience, science, beijing, brain, jij, retaining, arrival, connection, graded, cognitive, unstable, silence, prolonged, depression, poisson, marginally, dynamical] [working, rate, active, hold, critical, computational, theoretical, example, key, plateau, strong, solving]
On the relations of LFPs & Neural Spike Trains
David E. Carlson, Jana Schaich Borg, Kafui Dzirasa, Lawrence Carin
David E. Carlson, Jana Schaich Borg, Kafui Dzirasa, Lawrence Carin
[region, number, set, experiment, factor] [result, action, best, lower] [multiple, proposed, convolutional, network, novel, represented, predict, performance] [model, data, figure, inference, variational, process, local, distribution, predictive, bayesian, log, prior, band] [dictionary, clustering, element, analysis, detected] [lfp, rfe, cluster, sleep, cell, brain, time, single, predicting, dynamic, hippocampus, neuron, frequency, spike, animal, mouse, neural, vta, accumbens, thalamus, lfps, connectivity, individual, activity, awake, bin, nucleus, represent, recorded, akb, cortex, signal, change, dirichlet, djb, spiking, state, nature, shape, evolution, observed, contribution, neuronal] [functional, min, example]
Compressive Sensing of Signals from a GMM with Sparse Precision Matrices
Jianbo Yang, Xuejun Liao, Minhua Chen, Lawrence Carin
Jianbo Yang, Xuejun Liao, Minhua Chen, Lawrence Carin
[number, graphical, set, random, mrf] [unknown, algorithm, bound, gap, assuming, optimal] [reconstruction, proposed, image, ieee, training, pixel, based, performance, video, patch, frame, learning, level, neighborhood] [precision, gmm, gaussian, bayesian, prior, mixture, distribution, covariance, variational, gmrf, model, log, likelihood, incomplete, conditional, twist, markov, figure, data, measurement, ple, inference, estimate, posterior, true, imagery, constituted] [matrix, sensing, compressive, sparsity, sparse, denote, laplacian, analysis, estimating] [signal, drawn, reconstructed, hierarchical, estimated, simulated, inverse] [theoretical, linear, error, theorem, method, group, shrinkage, assume, lasso, empirical, estimation, statistical]
Finding a sparse vector in a subspace: Linear sparsity using alternating directions
Qing Qu, Ju Sun, John Wright
Qing Qu, Ju Sun, John Wright
[random, high, relaxation, rounding, arbitrary, consider, apply, experiment, solution, programming, set] [algorithm, phase, main, proof, space, result, bound, transition, conference] [learning, well, produce, proposed, based, work, face, input] [global, model, data, target, log] [sparse, vector, planted, subspace, adm, dictionary, alternating, arxiv, basis, preprint, analysis, nonzero, provided, row, matrix, nonconvex, sparsest, formulation, initialization, dimension, recovers, point, barak, recover, optimizer, sparsity, approximation, component, provably, recovery, element, recovering, succeeds, exactly, nonzeros, principal, etrapalli, overcomplete, close] [choice] [linear, problem, min, simple, direction, practical, method, theory, solve, theorem, biased, assume, large, theoretical, complexity, machine, suppose, guaranteed]
Online Optimization for Max-Norm Regularization
Jie Shen, Huan Xu, Ping Li
Jie Shen, Huan Xu, Ping Li
[number, collaborative, claim, corollary, high, paper, set] [online, algorithm, proposition, optimal, main, upper, proof, bounded, difference, access, studied] [learning, work, surrogate, pcp, proposed] [data, stationary, figure, limit, sample, third, converge] [matrix, uniformly, basis, norm, nuclear, rank, corruption, omrmd, factorization, pca, verify, component, dimension, low, subspace, analysis, sparse, decomposition, huan, mrmd, point, robust, equivalent, compute, row, surely, principal, completion] [produced, observed, underlying, fraction, memory] [problem, converges, theorem, regularized, prove, regularization, min, function, nathan, convergence, large, constrained, convex, optimization, machine, loss, stochastic, positive, solving, gradient, lipschitz, empirically, batch, optimize, journal, empirical, method, theoretical, term]
A Dual Algorithm for Olfactory Computation in the Locust Brain
Sina Tootoonian, Mate Lengyel
Sina Tootoonian, Mate Lengyel
[solution, mapping, number, random, binary, map, set] [feedback, environment, space, future] [learning, performance, activation, output, dense, body, work, propose, input] [full, model, figure, inference, log, concentration, observation, local, generative, posterior] [vector, matrix, small, sparse, projection, corresponding, component, read] [odor, pns, olfactory, reduced, lns, connectivity, circuit, antennal, readout, system, locust, lobe, noise, feedforward, mushroom, biological, ica, response, activity, neuron, glomerular, time, molecular, direct, change, eventually, observed, connection, reciprocal] [dual, problem, dimensional, form, lagrangian, consistent, gradient, note, positive, computational, require, descent, update, function, convergence]
Robust Tensor Decomposition with Gross Corruption
Quanquan Gu, Huan Gui, Jiawei Han
Quanquan Gu, Huan Gui, Jiawei Han
[solution, consider, high, corollary] [case, order, unknown, bound, lemma, previous, general, algorithm, study, goal, result, easy, bounded, deterministic, optimal, main] [well, work, performance, based, proposed, ieee] [observation, true, model, figure, variance, log, data, gaussian, restricted, estimate] [tensor, matrix, decomposition, corruption, norm, rank, robust, sparse, gross, noisy, analysis, operator, special, recover, singular, minimization, support, siam, randomly, trace, completion, decomposibility, component, residual, overlapped, unfolding, alternating, contaminated, condition] [noise] [error, convex, problem, assumption, theorem, estimation, theoretical, statistical, note, linear, size, min, regularization, optimization, solving, solve, large, method]
Tight convex relaxations for sparse matrix factorization
Emile Richard, Guillaume R. Obozinski, Jean-Philippe Vert
Emile Richard, Guillaume R. Obozinski, Jean-Philippe Vert
[set, number, sum, solution, equal, consider, relaxation, strength, random] [algorithm, upper, bound, oracle, order, case, lower, lemma, bounded, design] [propose, table, report, learning, multiple, based, denoising] [estimate, log, covariance, sample, sequential, figure, nmse] [norm, matrix, sparse, dimension, trace, rank, pca, factorization, small, atomic, low, component, formulation, vector, principal, compare, subspace, analysis, nonzero, decomposition, compute, consists, smaller, recovery, introduced, close] [magnitude, noise, varying] [statistical, convex, problem, min, linear, estimation, solve, form, computational, error, working, empirical, optimization, proximal, theoretical, assumption]
Discriminative Unsupervised Feature Learning with Convolutional Neural Networks
Alexey Dosovitskiy, Jost Tobias Springenberg, Martin Riedmiller, Thomas Brox
Alexey Dosovitskiy, Jost Tobias Springenberg, Martin Riedmiller, Thomas Brox
[number, set, random] [current, previous, algorithm] [feature, training, surrogate, learning, representation, convolutional, network, unsupervised, image, contrast, invariance, applied, invariant, accuracy, patch, performance, supervised, learned, discriminative, color, layer, input, learn, trained, visual, transformed, labeled, work, object, cnn, dataset, datasets, task, well, test, matching, recognition, spatial, pixel, softmax, hue, stl, extracted, based, deep, transformation, svm, denoising, pooling] [data, approach, unlabeled, average, log, sample, figure, standard, distribution] [analysis, vector, compare, distance, randomly, small] [neural, change, classification, magnitude] [class, objective, large, method, function, applying, problem]
Modeling Deep Temporal Dependencies with Recurrent Grammar Cells""
Vincent Michalski, Roland Memisevic, Kishore Konda
Vincent Michalski, Roland Memisevic, Kishore Konda
[mapping, set, structure, making, number] [sequence, conference, order, case, explicit, viewed] [learning, training, pgp, transformation, layer, hidden, test, frame, input, work, learn, representation, reconstruction, contrast, deep, trained, natural, rotation, onst, multiple, invariant, gae, feature, ieee, accuracy, higher, image, bilinear, predict, better, gating, bouncing, boltzmann, constancy] [model, prediction, data, predictive, figure, modeling, observation, standard] [relational, compute, orthogonal, seed] [time, neural, recurrent, represent, signal, capture, inferred, temporal, content, evolution] [machine, error, linear, acc, accelerated, constant, assumption, rate]
Recurrent Models of Visual Attention
Volodymyr Mnih, Nicolas Heess, Alex Graves, koray kavukcuoglu
Volodymyr Mnih, Nicolas Heess, Alex Graves, koray kavukcuoglu
[number, random, core, connected, ball, region] [agent, environment, policy, action, reward, game, decision, lower, case, static, deploy, maximize, search] [network, image, glimpse, object, attention, convolutional, mnist, location, visual, learn, trained, translated, ram, input, learning, representation, task, cluttered, training, detection, test, layer, scale, retina, scene, eye, table, resolution, learned, video, computer, output, train, hidden, outperforms, performance, vision, screen, baseline, combine] [model, processing, clutter, computation, allows, fully, full, figure, amount, distribution, log, approach] [vector] [neural, state, time, dynamic, control, recurrent, internal, hiddens] [sensor, gradient, size, error, linear, computational, simple]
Unsupervised learning of an efficient short-term memory network
Pietro Vertechi, Wieland Brendel, Christian K. Machens
Pietro Vertechi, Wieland Brendel, Christian K. Machens
[differential, set, apply, remains, balanced, random] [optimal, previous, history, minimize, equilibrium, case, follow, minimizes, order, chosen] [learning, input, network, coding, based, autoencoder, learn, work, output, reconstruction, unsupervised, performance, applied] [local, average, equation, approach, target, computation, step, data, reservoir] [matrix, minimization, fast, link, small, point] [recurrent, feedforward, time, neural, signal, spiking, rule, change, slower, centered, delayed, decoder, memory, represent, presynaptic, synaptic, actual, direct, white, remember, fxt, slow, dimensionality, biological, postsynaptic, reconstructed, derived, dynamical, operate, lateral] [loss, function, rate, note, term, cost, assume, linear, computational, relative, positive, respect, generally]
Optimal decision-making with time-varying evidence reliability
Jan Drugowitsch, Ruben Moreno-Bote, Alexandre Pouget
Jan Drugowitsch, Ruben Moreno-Bote, Alexandre Pouget
[belief, remains, solution, penalty, interval, high, making, programming] [optimal, reliability, bound, evidence, decision, current, policy, diffusion, expected, reward, momentary, accumulation, return, accumulate, associated, future, cir, accumulating, lower, michael, previous, deviation, implies, max, space] [performance, task, work, based, higher, well, illustrated, unit, feature, correct] [process, figure, model, larger, standard, distribution, expectation, approach, computation, equation, university] [requires, smaller, match, comparison, approximation] [time, single, dynamic, state, speed, trajectory, trial, change, independent, choice, perceptual, sensory] [constant, cost, depends, function, example, loss, large, method, journal, stochastic, linear, iteration, rate, size]
Optimal Teaching for Limited-Capacity Human Learners
Kaustubh R. Patil, Xiaojin Zhu, Łukasz Kopeć, Bradley C. Love
Kaustubh R. Patil, Xiaojin Zhu, Łukasz Kopeć, Bradley C. Love
[set, random, limited, item, provide, label, number, procedure, making, differ] [optimal, decision, spread, learner, order, study, feedback, normative, space] [training, teaching, test, human, gcm, category, teacher, capacity, performance, learning, idealization, work, exemplar, inconsistency, love, retrieved, similarity, trained, better, retrieval, differed, novel, performed] [model, figure, perform, distribution, true, parameter, conditional, equation, evaluate, university] [condition, small, distance] [memory, idealized, stimulus, response, cognitive, participant, individual, time, estimated, observed, experimental, neural] [machine, loss, function, stochastic, boundary, problem, error, optimization, bayes, journal, consistent, size, min]
Multi-View Perceptron: a Deep Model for Learning Face Identity and View Representations
Zhenyao Zhu, Ping Luo, Xiaogang Wang, Xiaoou Tang
Zhenyao Zhu, Ping Luo, Xiaogang Wang, Xiaoou Tang
[number, random, existing, continuous, facial, binary, set, remaining, experiment] [best, deterministic, lower, bound, setting, uniform, good] [face, view, identity, mvp, recognition, image, deep, learning, training, input, feature, hidden, viewpoint, output, primate, better, performance, testing, representation, compared, applied, network, ground, based, table, learned, convolutional, learn, multipie, fip, extract, reconstruction, discriminative, predict] [log, full, generate, data, model, figure, estimate, true, prior, sample, lbp, distribution] [spectrum, second, reconstruct, frontal, largest, recover, vector, unobserved] [reconstructed, neural, brain, raw, three, single, distinct] [perceptron, gradient, achieves, estimation, including]
Deep Joint Task Learning for Generic Object Extraction
Xiaolong Wang, Liliang Zhang, Liang Lin, Zhujin Liang, Wangmeng Zuo
Xiaolong Wang, Liliang Zhang, Liang Lin, Zhujin Liang, Wangmeng Zuo
[region, variable, set, discovery, binary, apply] [optimal, result, algorithm] [segmentation, object, network, learning, localization, joint, image, training, deep, salient, saliency, dataset, based, convolutional, detection, task, bounding, jointly, internet, reference, groundtruth, output, pixelwise, work, foreground, evaluation, testing, cpmc, applied, well, proposed, china, represented, indicates, ieee, propose, architecture, pattern, layer, compared, jaccard, selected, sliding, superpixels] [latent, model, data, full, approach, distribution, mask, estimate, sampling, mcmc, average, proposal, figure, sample] [compare, vector, iterative, requires, second, introduce] [neural, connection, time, presented] [method, term, cost, framework, optimize]
Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation
Emily L. Denton, Wojciech Zaremba, Joan Bruna, Yann LeCun, Rob Fergus
Emily L. Denton, Wojciech Zaremba, Joan Bruna, Yann LeCun, Rob Fergus
[number, biclustering, factor, connected, consider, structure] [gain, exploit, criterion, order] [convolutional, layer, performance, output, input, gpu, monochromatic, convolution, network, color, original, training, applied, spatial, feature, table, work, achieve, evaluation, compress, computed, computer, baseline, trained] [figure, approximated, prediction, fully, standard, covariance, approximate, computation, implementation] [approximation, cpu, svd, distance, decomposition, tensor, matrix, product, rank, second, speedup, preprint, compute, denote, arxiv, required, reduction, corresponding, redundancy, singular, clustering, small, greater, denotes, intermediate, improving] [neural, speed, described, memory, weight, cluster, presented] [empirical, large, outer, metric, loss, linear, size, compressing]
Unsupervised Deep Haar Scattering on Graphs
Xu Chen, Xiuyuan Cheng, Stephane Mallat
Xu Chen, Xiuyuan Cheng, Stephane Mallat
[connected, graph, set, partial, permutation, calculated, selection, maximum, random] [order, total, algorithm, unknown, selects, variation, appendix, minimizing, previous, grid] [scattering, haar, multiresolution, unsupervised, invariant, learning, transforms, scale, computed, computes, transform, deep, geometry, scrambled, supervised, image, wavelet, network, training, mnist, multiscale, feature, layer, percentage, pair, table, art, original, digit, representation, ieee, multiple, irregular, well] [data, figure, sampled, regression, fully, average] [absolute, compute, orthogonal, square, dimension, operator, approximation, sphere, basis, reduction, sparse, small, second, spectral, boolean] [signal, estimated, connectivity, state, neural] [error, size, applying, implemented, theorem, large, example]
Efficient Optimization for Average Precision SVM
Pritish Mohapatra, C.V. Jawahar, M. Pawan Kumar
Pritish Mohapatra, C.V. Jawahar, M. Pawan Kumar
[binary, set, discrete, number, sort, candidate] [algorithm, ranking, optimal, search, order, action, ranked, easy, upper] [svm, training, feature, object, score, input, output, joint, pascal, learning, proposed, dataset, bounding, selective, structured, detection, voc, table, select, propose, accuracy, retrieval, test, entire] [sample, inference, computation, approach, average, approximate, advantage, prediction] [rank, vector, compute, consists, exact, corresponding, second, matrix, computing, requires] [time, described, individual, neural, three, estimated] [negative, interleaving, loss, positive, optj, function, computational, complexity, violated, unimodal, framework, descending, optimizing, yue, domain, size, note, large, ssvm, appx, problem, method, class, require, ensuring, optimization, regularized, risk]
Ranking via Robust Binary Classification
Hyokun Yun, Parameswaran Raman, S. Vishwanathan
Hyokun Yun, Parameswaran Raman, S. Vishwanathan
[number, set, item, binary, collaborative, solution, quality, sum, construct, continuous, movie, user] [algorithm, ranking, context, bound, upper, explicit, relevant, easy, appendix, conference, minimize] [learning, performance, feature, retrieval, based, transformation, dataset, learn, evaluation, outperforms, score, datasets, better, training, multiple] [data, standard, model, parameter, latent, figure, parallel, computation, log] [robust, rank, top, vector, consists, ndcg] [time, song, independent, connection] [robirank, function, loss, objective, stochastic, gradient, optimization, convex, weston, machine, logistic, metric, note, problem, large, optimize, irpush, dcg, journal, infnormpush, distributed, statistical, linear, chapelle, basic, descent, ranksvm]
A* Sampling
Chris J. Maddison, Daniel Tarlow, Tom Minka
Chris J. Maddison, Daniel Tarlow, Tom Minka
[random, continuous, region, discrete, set, construction, number, energy, partition, node, maximum, interval, appear, perturbation, needed] [gumbel, bound, algorithm, max, upper, search, space, lower, difference, appears, case, perturbed, draw, gumbels, adaptive, probability, trick, optimal, exponentially] [computed, bounding, split, work, adding] [sampling, sample, process, likelihood, log, distribution, sampled, posterior, heap, data, drawing, density, model, gaussian, global, analogous, fewer, proposal, regression, sampler, allows, probabilistic] [exact, point, dimension, compute, second] [independent, noise, choice] [rejection, problem, function, argmax, cost, method, measure, linear, distributed, optimization, stochastic, note]
Minimax-optimal Inference from Partial Rankings
Bruce Hajek, Sewoong Oh, Jiaming Xu
Bruce Hajek, Sewoong Oh, Jiaming Xu
[random, item, partial, assignment, graph, high, number, independently, user, corollary, cdf, set, exists, accurately, apply, assign] [bound, lower, ranking, upper, probability, oracle, case, utility, general, chosen, thurstone, algorithm, proof, lemma, study, rescaled, logarithmic, main, break] [based, performance, select, work, hidden] [log, model, generated, scheme, full, likelihood, inference, parameter, mse, estimate, larger, true, infer] [pairwise, rank, denote, comparison, vector, uniformly, special, kmax, spectral, matrix] [breaking, preference, underlying, observed, independent, choice, derived] [theorem, estimator, error, positive, constant, estimation, suppose, function, inequality, assume, weighted, size]
Learning Time-Varying Coverage Functions
Nan Du, Yingyu Liang, Maria-Florina F. Balcan, Le Song
Nan Du, Yingyu Liang, Maria-Florina F. Balcan, Le Song
[set, random, number, maximum, covered, discrete, collection, continuous] [coverage, counting, diffusion, overage, earner, algorithm, cic, dic, social, mae, probability, combinatorial, martingale, bound, cumulative, tcoveragelearner, transmission, lemma, hellinger, parametrization, game, economics, customer, choose, submodular, best] [learning, based, learn, cascade, tij, performance, novel, selected, propose, multiple] [model, kernel, process, data, source, sample, likelihood, figure, distribution, estimate, regression, exponential, average, true, step, observation] [ridge, approximation, small, distance, dimension, synthetic, pairwise] [time, intensity, temporal, real, change, independent, estimated, event] [function, estimation, problem, optimization, theory, error, complexity, weighted, empirical, convex, method]
Online and Stochastic Gradient Methods for Non-decomposable Loss Functions
Purushottam Kar, Harikrishna Narasimhan, Prateek Jain
Purushottam Kar, Harikrishna Narasimhan, Prateek Jain
[set, cutting, penalty, partial, maximization, number, label] [online, regret, algorithm, appendix, uniform, lemma, result, design, proof, ranked, conference, bound, offer, case] [learning, performance, training, surrogate, datasets, well, stream, accuracy, proposed, svm, novel, achieve, table, task, scale] [data, average, pass, popular, figure, step, model, prediction] [point, top] [time, auc, length] [loss, pauc, method, gradient, stochastic, plane, structural, convex, function, framework, measure, large, arg, batch, buffer, epoch, theorem, convergence, positive, machine, risk, empirical, family, kdd, lipschitz, optimizing, size, require, min, note, suppose, cup, psg, hinge, roc, rate]
Tight Bounds for Influence in Diffusion Networks and Application to Bond Percolation and Epidemiology
Remi Lemonnier, Kevin Scaman, Nicolas Vayatis
Remi Lemonnier, Kevin Scaman, Nicolas Vayatis
[random, set, connected, edge, node, radius, pij, infection, hazard, infected, graph, corollary, degree, nyi, number, solution, maximization, independently, undirected, attachment, preferential, sigkdd, discovery, equal, tight, contagion, acm, subgraph, consider, existing, apply, spanning] [bound, proposition, expected, case, upper, order, percolation, probability, epidemiology, result, conference, transmission, diffusion, social, giant, lemma, exp, initial, regime, bond, general, difference] [network, cascade] [model, international, process, data, distribution, outgoing] [matrix, spectral, component, knowledge, totally, close, generalize] [behavior, time, independent, three, derived, inhomogeneous, removed] [size, derive, geometric, problem, rate, theory, note, stochastic, theoretical, hold]
Shaping Social Activity by Incentivizing Users
Mehrdad Farajtabar, Nan Du, Manuel Gomez-Rodriguez, Isabel Valera, Hongyuan Zha, Le Song
Mehrdad Farajtabar, Nan Du, Manuel Gomez-Rodriguez, Isabel Valera, Hongyuan Zha, Le Song
[user, number, maximization, sum, relation, set, structure, node, provide] [social, algorithm, appendix, branching, expected, budget, minimax, total, difference, history, minimum, counting, order] [network, based, evaluation, level] [process, multivariate, model, data, target, average, exponential, figure, perform, series, estimate] [matrix, rank, computing, compute, row, vector, second, product, nonnegative, sparse] [activity, intensity, event, exogenous, shaping, endogenous, hawkes, time, generation, three, poisson, service, independent, simulated, logarithm, correlation, system, trigger, real, capped, desired, drive, recurrent, roughly, usage, manuel, hongyuan, deg, cam] [framework, linear, objective, convex, method, gradient, optimization, theorem, form, solving, theoretical, summarizes, active, regularization, derive]
Difference of Convex Functions Programming for Reinforcement Learning
Bilal Piot, Matthieu Geist, Olivier Pietquin
Bilal Piot, Matthieu Geist, Olivier Pietquin
[number, programming, set, continuous, experiment, provide] [policy, optimal, mdps, obr, minimizing, space, obrm, order, api, explicit, couple, sequence, bound, avi, minimize, bellman, choose, mdp, reinforcement, algorithm, literature, action, controlling, criterion, probability, qapi, lower, horizon, garnet, called, proof, interesting, concentrability, qavi, case, reward, decision] [learning, representation, performance, better] [approximate, approach, allows, markov, sample, standard, variance, supplementary, distribution] [dca, norm, minimization, greedy, residual, consists, decomposition, compute, approximation, second] [state, dynamic, choice, real, intermediary] [function, convex, term, problem, empirical, large, error, respect, optimization, iteration, linear, estimation, solve, batch, solved, theorem]
On Communication Cost of Distributed Statistical Estimation and Dimensionality
Ankit Garg, Tengyu Ma, Huy Nguyen
Ankit Garg, Tengyu Ma, Huy Nguyen
[private, random, factor, message, sum, high, number, consider, public, provide, needed, center, independently] [lower, bound, minimax, proof, case, upper, study, unknown, lemma, general, main, setting] [achieve, task, learning, better, question] [distribution, gaussian, sample, parameter, data, log, amount, estimate, standard, prior] [simultaneous, sparse, dimension, ith, improved, estimating, denote, technique] [direct, dimensionality] [communication, cost, protocol, theorem, loss, problem, machine, distributed, estimation, squared, dimensional, complexity, statistical, interactive, note, prove, coordinate, estimator, solving, simple, tradeoff, solves, error, send, randomness, proving, transcript, conjecture, denoted, essentially, suppose, rpub, icv, applying]
How hard is my MDP?" The distribution-norm to the rescue"
Odalric-Ambrym Maillard, Timothy A. Mann, Shie Mannor
Odalric-Ambrym Maillard, Timothy A. Mann, Shie Mannor
[number, consider, notion, existing, apply, factor, provide] [mdps, bound, mdp, reinforcement, policy, transition, regret, algorithm, weissman, environmental, ucrl, lemma, undiscounted, conference, discounted, probability, proposition, benchmark, optimal, result, space, vmax, diameter, jaksch, case, order, decision, proof] [learning, based, proposed, natural, work, better] [kernel, distribution, concentration, sample, international, markov, model, mass] [norm, analysis, small, corresponding, denote, smaller, low, technical] [state, dependent, common, capture, estimated, typical] [hardness, function, respect, inequality, measure, theorem, machine, empirical, error, dual, quantity, complexity, metric, inf, theoretical]
Conditional Swap Regret and Conditional Correlated Equilibrium
Mehryar Mohri, Scott Yang
Mehryar Mohri, Scott Yang
[exists, notion, consider, remark, random, number, partial, map] [regret, swap, player, algorithm, action, correlated, minimizing, competitor, online, equilibrium, game, bound, bigram, scenario, bandit, sequence, repeated, bounded, follow, proof, played, previous, sublinear, appendix, round, call, result, lmin, history, transducer, general, oblivious, extension, case, chooses, write, choose, admits] [learning, joint, natural, work] [conditional, distribution, standard, log, figure, stationary] [denote, minimization, vector, provided, matrix, power, introduced, interpreted, introduce] [time, external, achieving, interpretation] [loss, class, min, theorem, empirical, function, form, randomized, computational, complexity, derive, converges, inequality, weighted]
RAAM: The Benefits of Robustness in Approximating Aggregated MDPs in Reinforcement Learning
Marek Petrik, Dharmashankar Subramanian
Marek Petrik, Dharmashankar Subramanian
[set, robustness, solution, constructed, program, reduce, consider, cplex, programming, factor] [aggregation, policy, raam, optimal, mdp, algorithm, rmdp, return, transition, bound, decision, aggregated, occupancy, mdps, reward, main, conference, bounded, rmdps, general, van, upper, proof, lemma, uniform, action, wiesemann, result] [performance, learning, based, original, improve] [uncertainty, approximate, markov, concentration, standard, aggregate, international, distribution, full, figure, parameter, approach] [robust, approximation, denote, comparison, compute, norm] [state, dynamic, represent, time, frequency, choice] [function, loss, linear, theorem, example, computational, error, solving, machine, dual, min, large, note, regular, assume, stochastic, structural]
Optimizing Energy Production Using Policy Search and Predictive State Representations
Yuri Grinberg, Doina Precup, Michel Gendreau
Yuri Grinberg, Doina Precup, Michel Gendreau
[constraint, solution, set, energy, equal, structure, programming, number] [policy, search, algorithm, future, current, easy, total, space, ithin, initial, period, expected] [based, representation, complex, learning, applied, well, proposed, improvement, produce] [water, model, predictive, amount, turbine, reservoir, approach, week, site, pred, hydroelectric, generative, data, production, process, parameter, estimate, historical, electricity, figure, bec, headwater, psr, volume, psrs, href, plant, developed, modeling, modelling, renewable, earch] [power, comparison, knowledge] [state, speed, time, dynamic, control, three, produced, change, quantitative, represents, simulator] [problem, optimization, stochastic, domain, linear, optimizing, descent]
Near-optimal Reinforcement Learning in Factored MDPs
Ian Osband, Benjamin Van Roy
Ian Osband, Benjamin Van Roy
[structure, set, high, consider, graph, random, polynomial, number, mapping, sum, shorthand, apply] [factored, mdp, regret, reinforcement, probability, bound, policy, optimal, psrl, algorithm, reward, mdps, transition, cient, action, write, result, agent, lemma, fmdp, exponentially, planner, conference, expected, optimistic, sequence, environment, michael, upper, ronald, daphne, exploit, erence, carlos, setting, deviation, bounded, follow, proof, diameter, deterministic, appendix, adaptive] [learning, performance] [log, approximate, prior, sample, posterior, distribution, sampling, international, true, markov, fully] [analysis, span, introduce] [time, state, dynamic, dimensionality] [function, machine, problem, empirical, theorem, large, class, simple, practical, term]
Learning Neural Network Policies with Guided Policy Search under Unknown Dynamics
Sergey Levine, Pieter Abbeel
Sergey Levine, Pieter Abbeel
[constraint, entropy, arbitrary, partially, programming, high] [policy, search, algorithm, guided, conference, octopus, itr, unknown, minimize, arm, expected, hole, swimming, general, current, initial] [learning, learn, network, complex, learned, trained, work, training, task, well] [prior, distribution, international, gaussian, model, local, approach, gps, target, gmm, fewer, sample, covariance, step] [distance, required] [trajectory, neural, parameterized, time, peg, state, observed, dynamic, insertion, controller, rwr, system, ilqg, pilco, cem, optimized] [method, linear, cost, optimization, dual, respect, optimize, quadratic, objective, optimizes, machine, constrained, lagrangian, require, problem]
Optimizing F-Measures by Cost-Sensitive Classification
Shameem Puthiya Parambath, Nicolas Usunier, Yves Grandvalet
Shameem Puthiya Parambath, Nicolas Usunier, Yves Grandvalet
[binary, set, maximization, label, asymmetric] [optimal, proposition, probability, result, decision, algorithm, goal, space, general, proof, case] [false, learning, svm, performance, multiclass, level, svms, dataset, datasets, training, based] [distribution, approach, volume, posterior, prediction, regression, international, sample, data] [thresholding, reduction, analysis, vector, approximation] [presented] [multilabel, class, function, optimizing, positive, threshold, problem, error, svmperf, logistic, measure, thresholded, cost, linear, negative, loss, optimize, practical, assume, theoretical, machine, framework, denoted, generalization, written, corresponds, method, asymmetry, example, inequality, large, note, respect, theorem, empirical]
Analysis of Learning from Positive and Unlabeled Data
Marthinus C. du Plessis, Gang Niu, Masashi Sugiyama
Marthinus C. du Plessis, Gang Niu, Masashi Sugiyama
[penalty, label, consider] [optimal, decision, conference, minimize, order, expected, minimizes, unknown, difference, probability, algorithm] [learning, dataset, false, supervised, accuracy, detection, illustrated, labeled, training, surrogate, table, train, performed] [prior, unlabeled, data, true, figure, fully, international, lead] [outlier, analysis, supported, worse, vector, support, analyze, denote] [estimated, tokyo, modulation] [class, loss, positive, hinge, risk, ramp, function, error, problem, rate, negative, threshold, ordinary, effective, generalization, estimation, boundary, excess, large, term, objective, empirical, convex, solved, practical, machine, note, practice, weighted, depends, constant, usps, theoretical]
Deconvolution of High Dimensional Mixtures via Boosting, with Application to Diffusion-Weighted MRI of Human Brain
Charles Y. Zheng, Franco Pestilli, Ariel Rokem
Charles Y. Zheng, Franco Pestilli, Ariel Rokem
[set, penalty, solution, number, consider, path, continuous] [algorithm, diffusion, proposition, oracle, minimum, best, learner, minimize] [boost, deconvolution, applied, transformed, ensemble, multiple, accurate, input] [mixture, model, dwi, kernel, ebp, nnls, data, global, figure, parameter, discretization, true, crossing, voxel, distribution, demonstrate, fascicle, prediction, fodf, simulation, approach, nerve, estimate, mri, measurement] [basis, residual, elastic, pursuit, tensor, cbp, totally, refer] [signal, brain, single, estimated, imaging, white, orientation] [function, boosting, active, iteration, problem, error, weak, regularization, direction, convergence, size, objective, descent, assume, stochastic, family, dti, estimation, computational]
Bayesian Nonlinear Support Vector Machines and Discriminative Factor Modeling
Ricardo Henao, Xin Yuan, Lawrence Carin
Ricardo Henao, Xin Yuan, Lawrence Carin
[factor, set, expression, attractive, maximization, number] [decision, previous, general, case] [svm, discriminative, well, learning, table, work, feature, joint, scale, proposed, svms, better, dataset, place] [bayesian, model, prior, data, inference, distribution, gaussian, conditional, covariance, gene, mcmc, posterior, likelihood, perform, latent, mixture, ndfm, probit, process, ecm, standard, sampling, modeling, laplace, exponential, density, sampler, log, approximate, bsvm, ldfm, straightforward, markov, allows, elliptical, fitc, developed, predictive] [vector, matrix, support, sparse, slice, link, formulation, approximation] [nonlinear, directly, cell, three, hierarchical] [linear, function, error, loss, margin, boundary, hinge, large, usps, implemented, form, note, corresponds, regularization, complexity, term]
On the Statistical Consistency of Plug-in Classifiers for Non-decomposable Performance Measures
Harikrishna Narasimhan, Rohit Vaish, Shivani Agarwal
Harikrishna Narasimhan, Rohit Vaish, Shivani Agarwal
[continuous, set, apply, binary, consider, monotonically, number, monotonic, exists] [algorithm, probability, optimal, result, regret, lemma, cpe, considered, proof, general, main, maximize, case] [performance, training, learning, work, learn, test, template, text, table] [distribution, data, true, sample, estimate, figure, expectation] [suitable, thresholding, synthetic, support] [expressed, real, drawn, underlying] [class, consistency, assumption, tpr, measure, threshold, empirical, function, bayes, svmperf, method, theorem, tnr, form, thresholded, consistent, rate, increasing, tnrd, statistical, requisite, imbalanced, optimizing, continuity, tprd, optimize, convergence, risk, sign, noting, theoretical, ptprd, note, applying, argmax, geometric, journal]
Exponential Concentration of a Density Functional Estimator
Shashank Singh, Barnabas Poczos
Shashank Singh, Barnabas Poczos
[entropy, random, consider, continuous, exists] [bound, conference, bounded, exp, case, general, lower, proof, main, probability, bandwidth, lemma, extend] [learning, based, work, ieee, test, feature, image, testing] [density, conditional, concentration, bias, exponential, kernel, nonparametric, estimate, sample, data, variance, integral, international, distribution, processing, kde, university, independence] [estimating, denotes, denote, vector, condition, notation] [mutual, independent, state, triangle] [inequality, estimator, functional, functionals, estimation, machine, divergence, mirrored, cmi, rate, class, problem, convergence, applying, dependence, including, nhd, error, constant, assume, prove, dxi, kdj, proven, suppose, function, squared, theoretical, lipschitz, statistical, derive, converges, note, distributed, depends]
Altitude Training: Strong Bounds for Single-Layer Dropout
Stefan Wager, William Fithian, Sida Wang, Percy S. Liang
Stefan Wager, William Fithian, Sida Wang, Percy S. Liang
[naive, label, def, number, red, set, random] [bound, decision, probability, conference, result, good, proposition, adaptive, best, michael, upper, appears, goal, optimal, minimizer] [dropout, training, err, learning, test, original, word, long, trained, discriminative, feature, score, improves, well, work, natural, language, sentiment, association] [generative, model, figure, regression, international, bias, data, average, larger, processing, full] [analysis, corrupted, vector, small] [topic, poisson, document, neural, binomial, sida, drawn, rare] [error, risk, bayes, generalization, excess, linear, logistic, theorem, assumption, rate, measure, machine, erm, boundary, empirical, large, altitude, assume, class, size, note, suppose, loss, theoretical, regularization, term]
Simultaneous Model Selection and Optimization through Parameter-free Stochastic Learning
Francesco Orabona
Francesco Orabona
[set, solution, number, corollary, high, selection, paper, consider] [optimal, algorithm, online, bound, regret, case, best, setting, iid, probability, result, oco, space, competitor, lemma] [learning, performance, training, svm, work, proposed, train] [step, kernel, model, processing, prior, standard] [vector, square, approximation, discussion, knowledge, norm, proved, condition] [neural, range, time, independent] [stochastic, convergence, pistol, loss, averaged, regularization, rate, note, theorem, function, assume, gradient, inf, risk, size, descent, convex, form, excess, statistical, problem, predictor, linear, sgd, asgd, prove, theoretical, practical, machine, require, dimensional, batch, optimization, regularized, smooth, method, perceptron, lipschitz, theory]
Positive Curvature and Hamiltonian Monte Carlo
Christof Seiler, Simon Rubinstein-Salzedo, Susan Holmes
Christof Seiler, Simon Rubinstein-Salzedo, Susan Holmes
[high, number, paper, remark, geodesic, notion, random, potential] [space, probability, case, order, lip, bound, starting, minimum, choose, main, result] [manifold, coarse, work, geometry, long] [curvature, riemannian, sample, hmc, sectional, qqq, jacobi, concentration, hamiltonian, markov, monte, chain, distribution, multivariate, gaussian, carlo, sampling, tangent, standard, figure, mcmc, ricci, bayesian, density, grad, joulin, secx, ollivier, anatomy, expectation, step] [running, dimension, point, euclidean, distance, numerical, sphere] [state, inverse, time, three, length] [metric, positive, inequality, error, theoretical, dimensional, computational, inf, measure, calculate, problem, theorem, function, note, depends, constant]
Bayes-Adaptive Simulation-based Search with Value Function Approximation
Arthur Guez, Nicolas Heess, David Silver, Peter Dayan
Arthur Guez, Nicolas Heess, David Silver, Peter Dayan
[belief, continuous, root, discrete, number, set, consider, balancing, tree, solution, provide, map, procedure] [planning, search, algorithm, history, pendulum, bamcp, bafa, current, policy, decision, agent, optimal, conference, return, space, action, reward, reinforcement, probability, environment, version, future, bandit, height, associated, general, horizon, order, online, result, discounted, starting] [learning, representation, based, applied] [sampling, figure, approach, simulation, step, bayesian, international, model, parametric, prior, gaussian, uncertainty, distribution, posterior, monte, markov, carlo] [approximation, small, generalize, vector] [state, control, interaction, time, navigation] [function, problem, generalization, large, method, machine, form, updated]
Fast Multivariate Spatio-temporal Analysis via Low Rank Tensor Learning
Mohammad Taha Bahadori, Qi (Rose) Yu, Yan Liu
Mohammad Taha Bahadori, Qi (Rose) Yu, Yan Liu
[solution, mode, structure, construct, constraint, existing] [algorithm, optimal, appendix, side] [learning, dataset, spatial, training, task, table, accurate, cross, datasets] [data, multivariate, model, global, local, gaussian, process, series, forward, covariance, rmse, computationally, regression, prediction, parameter, estimate, mixture, full, figure] [tensor, rank, greedy, cokriging, low, matrix, orthogonal, forecasting, laplacian, analysis, norm, ccds, decomposition, multitask, singular, ushcn, climate, synthetic, eigenvalue, yelp, running, fast, formulation, foursquare, kriging, tucker, friendship] [time, capture, three, shared, temporal] [estimation, problem, framework, admm, method, argmin, complexity, consistency, optimization, solve, loss, convex, form, solving]
Scalable Inference for Neuronal Connectivity from Calcium Imaging
Alyson K. Fletcher, Sundeep Rangan
Alyson K. Fletcher, Sundeep Rangan
[factor, set, number, passing, message, procedure, potential, random, node, graph, typically] [current, initial, total, associated] [hidden, network, proposed, computed, challenging, performed, based, frame, image] [estimate, parameter, standard, approximate, model, distribution, inference, step, computationally, log, computation, expectation, bayesian, likelihood, concentration] [matrix, amp, exact, vector] [calcium, time, imaging, connectivity, spike, neuron, neural, scalar, state, system, loopy, wij, dynamical, membrane, described, wsk, voltage, observed, estimated, nonlinear, dki, typical, inferring, neuronal, noise, integration] [estimation, linear, method, rate, key, function, constant, simple, optimization, computational, assume, negative, relative, regularization]
Semi-Separable Hamiltonian Monte Carlo for Inference in Bayesian Hierarchical Models
Yichuan Zhang, Charles Sutton
Yichuan Zhang, Charles Sutton
[energy, potential, consider, number] [call, previous, called, general, algorithm] [joint, manifold, table, learning] [hamiltonian, leapfrog, sshmc, monte, rmhmc, hmc, log, gibbs, carlo, step, abla, model, sample, hyperparameters, hyperparameter, mass, posterior, bayesian, distribution, gaussian, separable, kinetic, chain, latent, data, auxiliary, mcmc, variance, sampling, figure, fim, mixing, markov, allows, process, mix, riemannian, sampled, simulation, funnel, girolami, standard] [matrix, hessian, point, faster, alternating] [hierarchical, time, independent, neural, property] [method, statistical, computational, group, function, size, min, metric, constant, stochastic, machine, logistic, journal, cost, dimensional]
Universal Option Models
hengshuai yao, Csaba Szepesvari, Richard S. Sutton, Joseph Modayil, Shalabh Bhatnagar
hengshuai yao, Csaba Szepesvari, Richard S. Sutton, Joseph Modayil, Shalabh Bhatnagar
[solution, set, user, constructed, mapping, move, number, random] [option, reward, policy, uom, uoms, return, expected, probability, action, game, planning, reinforcement, best, universal, discounted, transition, mdp, loem, previous, associated, tabular, occupancy, terminal, initial, goal, strategy, termination] [learning, selected, unit, abstract, learned, select, feature, multiple, traditional, reference, computed, database] [model, distribution, process, step, figure, standard, approach] [approximation, vector, matrix, uniformly, introduce, provided, compute] [state, time, article, underlying, independent, science, temporal] [function, linear, theorem, query, assume, note, large, method, updated, prove, predicts]
Cone-Constrained Principal Component Analysis
Yash Deshpande, Andrea Montanari, Emile Richard
Yash Deshpande, Andrea Montanari, Emile Richard
[maximum, consider, witness, constraint, solution, message, exists, tight, provide, number, passing] [case, phase, minimax, bound, algorithm, optimal, unknown, upper, proof, result, maximize] [additional, ieee, retrieval, achieved] [likelihood, model, estimate, figure, data, gaussian, standard, observation, local] [cone, power, principal, sparse, matrix, vector, pcn, orthant, projection, analysis, provided, eigenvector, component, lim, pca, characterize, noisy, iterative, exact, eigenvalue, special, rank, arxiv, xvt, point, andrea, largest, initialization, preprint, amp] [signal, subject, noise] [problem, risk, theorem, estimator, convex, iteration, estimation, dual, prove, quadratic, polyhedral, empirical, asymptotic, positive, note, simple, optimization, theory, statistical, function, large, theoretical, solving, closed]
Bayesian Sampling Using Stochastic Gradient Thermostats
Nan Ding, Youhan Fang, Ryan Babbush, Changyou Chen, Robert D. Skeel, Hartmut Neven
Nan Ding, Youhan Fang, Ryan Babbush, Changyou Chen, Robert D. Skeel, Hartmut Neven
[energy, random, number, differential, potential, set] [unknown, appendix, general, algorithm, conference, main, total, probability, equilibrium] [test, correct, dataset, mnist, ensemble, based, learning, additional, produce] [distribution, sgnht, sampling, density, canonical, kinetic, sghmc, rmse, target, bayesian, discretization, monte, figure, step, thermostat, true, icml, variance, metropolis, force, log, hamiltonian, international, parameter, estimate, temperature, posterior, sample, plot, marginalization, generate, hmc, carlo, mcmc, university, minibatch, thermal] [small, condition, matrix, smaller] [noise, langevin, time, system, dynamic, recipe, perplexity, represents, dirichlet, independent] [stochastic, gradient, method, machine, error, theorem, large, statistical, size, normal, proper, function, constant, estimation]
Just-In-Time Learning for Fast and Flexible Inference
S. M. Ali Eslami, Daniel Tarlow, Pushmeet Kohli, John Winn
S. M. Ali Eslami, Daniel Tarlow, Pushmeet Kohli, John Winn
[factor, message, set, random, number, marginal, mapping, forest, leaf, typically, consider, mar] [oracle, context, general, daniel, decision] [learning, training, accuracy, output, input, learn, natural, predicted, represented, challenging, computed] [inference, jit, sampling, uncertainty, regression, yield, log, outgoing, model, incoming, gaussian, data, regressors, compound, regressor, distribution, prediction, aware, sample, figure, temperature, xout, precision, perform, moracle, consultation, umax, dkl, monte, carlo, highly, computationally, probabilistic, average, expectation, propagation, mout, whilst] [knowledge, compute, small] [speed, time, gamma, inferred, held] [logistic, rate, min, error, function, knn, framework, problem, note, require]
Incremental Local Gaussian Regression
Franziska Meier, Philipp Hennig, Stefan Schaal
Franziska Meier, Philipp Hennig, Stefan Schaal
[localized, set, number, high] [online, bound, algorithm, space, version] [learning, training, test, learn, input, joint, table, feature, performance, achieve, well, datasets, trained] [local, data, gaussian, model, regression, lgr, log, variational, lwpr, posterior, nmse, approximate, global, lwr, process, probabilistic, bayesian, stefan, prediction, inference, mse, sarcos, edward, fewer, incoming, evaluate, generative, kernel, international, hyperparameters, likelihood, robot] [incremental, sparse, point, distance, spectrum, vector, approximation, component] [neural, inverse, control, length, rhythmic, real, consistently] [function, weighted, locally, linear, batch, computational, large, error, form, update, cost, machine, normalized, gradient]
Communication Efficient Distributed Machine Learning with the Parameter Server
Mu Li, David G. Andersen, Alex J. Smola, Kai Yu
Mu Li, David G. Andersen, Alex J. Smola, Kai Yu
[worker, reduce, partition, synchronization, set, scheduler] [algorithm, pull, general, appendix, best, previous] [learning, task, scale, network, training, proposed, well, reconstruction, dataset, performance] [data, parameter, model, push, figure, regression, dependency, sequential, parallel, advantage, scalable, larger] [sparse, compute] [system, time, shared, delayed, change, range, generation] [server, distributed, machine, consistency, gradient, proximal, block, iteration, framework, optimization, convergence, large, key, rate, delay, solving, logistic, communication, updated, waiting, size, descent, assume, method, group, kkt, solve, function, update, subgradient, compressing, objective, caching, communicated, coordinate, petuum]
Distributed Bayesian Posterior Sampling via Moment Sharing
Minjie Xu, Balaji Lakshminarayanan, Yee Whye Teh, Jun Zhu, Bo Zhang
Minjie Xu, Balaji Lakshminarayanan, Yee Whye Teh, Jun Zhu, Bo Zhang
[number, node, factor, consider, subset] [algorithm, previous, good] [combination, dataset, natural, learning, ground, well, wang, truth, better, based, computed, collected] [posterior, local, mcmc, bayesian, parameter, data, prior, moment, approximate, exponential, sampling, figure, distribution, sampler, multivariate, gaussian, likelihood, sharing, sample, parallel, approximated, variational, slab, regression, density, global, computation, tested, scot, inference, standard, monte, david, estimate, covariance, conditional, processing, generated, carlo, larger, mse, embarrassingly] [approximation, denote, vector, product, sparse, compute, consists, smaller] [estimated, spike, drawn] [convergence, family, distributed, method, machine, stochastic, large, linear, note, iteration, logistic, asynchronous, journal, positive]
Communication-Efficient Distributed Dual Coordinate Ascent
Martin Jaggi, Virginia Smith, Martin Takac, Jonathan Terhorst, Sanjay Krishnan, Thomas Hofmann, Michael I. Jordan
Martin Jaggi, Virginia Smith, Martin Takac, Jonathan Terhorst, Sanjay Krishnan, Thomas Hofmann, Michael I. Jordan
[number, procedure, naive, paper, worker, making] [algorithm, case, online, best, order, setting, michael, berkeley] [work, learning, proposed, datasets, well] [local, data, parallel, log, parameter, processing, amount, figure, computation, average, approach, larger] [analysis, vector, smaller, corresponding] [single, time, ascent, internal, presented] [distributed, coordinate, dual, convergence, communication, method, primal, optimization, rate, descent, stochastic, suboptimality, ocal, sgd, loss, martin, machine, linear, size, batch, framework, communicated, peter, block, cocoa, inner, problem, ual, gradient, theoretical, example, objective, assume, large, outer, sdca, iteration, class, cov, locally, nathan, convex]
Multi-Step Stochastic ADMM in High Dimensions: Applications to Sparse Optimization and Matrix Decomposition
Hanie Sedghi, Anima Anandkumar, Edmond Jonckheere
Hanie Sedghi, Anima Anandkumar, Edmond Jonckheere
[high, consider, provide, number, constraint] [online, bound, optimal, algorithm, minimax, initial, version, lower, probability, setting, access] [table, multiple, better] [log, radar, impose, parameter, scaling, local, sample, step] [matrix, sparse, rank, decomposition, low, norm, alternating, sparsity, spectral, projection, square, arxiv, nuclear, preprint, noisy, comparison, vector, approximation, compare] [independent, noise, length] [method, convergence, rate, admm, reason, optimization, epoch, stochastic, error, function, loss, problem, strong, update, batch, assumption, convexity, regularization, linear, inexact, min, direction, arg, regularized, constant, note, dual, form, statistical, require, simple, dimensional, agarwal, theorem, alm, tttt]
Stochastic Proximal Gradient Descent with Acceleration Techniques
Atsushi Nitanda
Atsushi Nitanda
[set, number, random, consider, equal, paper, monotonically] [lemma, proof, sequence, appropriate, case, best, bound, history] [learning, proposed, table, better, zhang, performs, training, achieve, propose] [variance, log, figure, processing, comparable, expectation, allows, supplementary, popular] [reduction, component, introduced, technique, randomly, analysis, proximity] [system, neural, ascent, described] [gradient, stochastic, eik, method, complexity, proximal, descent, rate, size, accelerated, apg, convergence, assumption, iteration, dual, acceleration, theorem, achieves, regularized, linear, variant, convex, min, machine, coordinate, inequality, pgd, respect, regularization, spgd, note, solving, distributed, logistic, arg, suppose, prove, function, optimization, lipschitz, large, svrg, loss, objective]
Beyond the Birkhoff Polytope: Convex Relaxations for Vector Permutation Problems
Cong Han Lim, Stephen Wright
Cong Han Lim, Stephen Wright
[sorting, permutation, solution, birkhoff, polytope, ordering, permutahedron, set, relaxation, xin, rqps, gurobi, constraint, fogel, continuous, paper, number, tiebreaking, graph, award, penalty, phn] [algorithm, version, good, side, case, optimal, best] [network, similarity, additional, better, based, input, performance, describe, reported, computer] [figure, markov, sample, xout, log, chain, plot, covariance] [formulation, matrix, seriation, spectral, hull, vector, introduce, projection, point, refer, introduced, compact, symmetric, matlab, noisy, compare, recovers, noiseless, closer, fiedler, analysis] [time, noise, experimental] [convex, problem, objective, regularization, min, regularized, optimization, large, method, constant, solve]
A statistical model for tensor PCA
Emile Richard, Andrea Montanari
Emile Richard, Andrea Montanari
[random, consider, message, passing, ratio, maximum, high, solution] [case, lemma, order, bound, side, probability, upper, result, lower, sequence, algorithm, version, appears] [recursive, additional, based, natural] [model, standard, estimate, approximate, approach, likelihood, local, tractable, gaussian, processing, log, data] [tensor, matrix, pca, power, amp, lim, analysis, symmetric, unfolding, denote, matq, spiked, vector, principal, singular, component, suggests, norm, operator, simultaneous, establish, successful, sharp, surely, provided, condition, compressed] [noise, neural, state, signal, behavior, desired, correlation] [iteration, theorem, problem, large, statistical, theory, estimator, method, optimization, journal, estimation, simple, computational, converges, normal, loss, convex, assume, prove, complexity]
large scale canonical correlation analysis with iterative least squares
Yichao Lu, Dean P. Foster
Yichao Lu, Dean P. Foster
[number, experiment, random, solution, structure, set, consider] [algorithm, space, search, conference, case] [word, performance, dataset, based, learning, datasets, output, well, computed] [canonical, data, supplementary, extremely] [cca, top, fast, singular, kpc, kcca, ling, rpcca, matrix, huge, frequent, compute, cpu, decomposition, krpcca, projection, cxx, orthogonal, iterative, cyy, principle, sparse, introduce, analysis, column, detailed, small, approximation, cxy, ykcca, computing, exact, vector, diagonal, denote, xkcca, dean, introduced, low, ptb, consists, mentioned, svd] [correlation, three, slow, time] [large, gradient, iteration, theorem, problem, machine, dimensional, error, descent, solving, size]
Accelerated Mini-batch Randomized Block Coordinate Descent Method
Tuo Zhao, Mo Yu, Yiming Wang, Raman Arora, Han Liu
Tuo Zhao, Mo Yu, Yiming Wang, Raman Arora, Han Liu
[set, partial, number, existing, consider, solution, tmax, equal, loop, corollary] [algorithm, choose, probability, strategy, best, sequence, appendix, lemma, exploit] [performance, learning, proposed, based, selected, better] [variance, step, sample, scheme, sampled, figure, sampling, regression, approximate] [randomly, sparse, exact, component, attains, introduced, reduction, vector, support, compare, minimization] [estimated] [gradient, method, descent, block, iteration, mrbcd, stochastic, active, coordinate, proximal, size, regularization, computational, randomized, complexity, rbcd, empirical, spvrg, convergence, regularized, convex, risk, linear, brbcd, rfi, outer, optimization, inner, function, calculate, rgj, journal, positive, machine, assumption, rate, objective, argmin, large, note, lmax, tong]
Sparse PCA with Oracle Property
Quanquan Gu, Zhaoran Wang, Han Liu
Quanquan Gu, Zhaoran Wang, Han Liu
[high, penalty, set, relaxation, solution, existing, limited, concave, number] [oracle, algorithm, case, optimal, probability, previous, minimax] [based, proposed, dataset, datasets, novel] [covariance, true, sample, standard, log, parameter, local] [sparse, principal, matrix, subspace, projection, nonconvex, pca, support, spca, synthetic, fantope, eigenvectors, component, norm, corresponding, analysis, condition, sharper, recovery, leading, frobenius, rank, arxiv, mcp, preprint, recovers, attains, eigenvalue, exactly, sparsity, nonzero, suitable, introduce, top, recover, spiked, enjoys] [magnitude, population, property, correlation] [estimator, rate, convex, convergence, statistical, family, note, regularization, assumption, estimation, journal, annals, problem, error, achieves, optimization, theorem, large, method, prove, theoretical, term, positive, empirical]
Reputation-based Worker Filtering in Crowdsourcing
Srikanth Jagabathula, Lakshminarayanan Subramanian, Ashwin Venkataraman
Srikanth Jagabathula, Lakshminarayanan Subramanian, Ashwin Venkataraman
[worker, label, honest, reputation, adversarial, labeling, assignment, adversary, number, set, hard, soft, graph, random, crowdsourcing, penalty, consider, malicious, existing, provide, high, assigned, bipartite, identify, fact, quality, binary, degree, absence, assign, lij] [algorithm, aggregation, strategy, optimal, decision, case, conference, lower, probability, general, reliability, uniform, bound, goal, best, main, online] [task, performance, based, accuracy, labeled, well, presence, propose, higher, indicates, idea] [true, standard, sophisticated, model, international, perform, infer, probabilistic, average] [denote, provided, low, matrix] [rule, inferring] [theorem, note, theoretical, incorrect, class, suppose]
Extremal Mechanisms for Local Differential Privacy
Peter Kairouz, Sewoong Oh, Pramod Viswanath
Peter Kairouz, Sewoong Oh, Pramod Viswanath
[privacy, mechanism, differential, staircase, private, differentially, binary, privatization, induced, marginals, high, alphabet, program, provide, maximization, analyst, extremal, exists, privatized, symposium, consider, converse, interested, solution, maximizes, geng, fundamental] [optimal, utility, bound, theoretic, maximize, maximizing, optimality, setting, total, type, order, probability, annual, case, general, follow, study, achievable] [pair, output, testing, representation, computer, input] [data, local, dkl, distribution, generated, global, figure, sample] [low, arxiv, preprint, equivalent, matrix, denote] [response, nonlinear, choice] [statistical, theorem, randomized, linear, hypothesis, problem, error, positive, family, solving, simple, theory, depends, convex, function, tradeoff, divergence, prove]
The Large Margin Mechanism for Differentially Private Maximization
Kamalika Chaudhuri, Daniel J. Hsu, Shuang Song
Kamalika Chaudhuri, Daniel J. Hsu, Shuang Song
[private, differential, privacy, differentially, maximization, mechanism, itemset, mining, item, set, universe, number, high, kamalika, notion, maximum, consider, provide, shell, adam, maximizers, applicability, kobbi, procedure, itemsets, abhradeep, chaudhuri, cynthia] [algorithm, utility, probability, bound, general, goal, lower, highest, design, guarantee, pure, order, clear, case, previous] [learning, dataset, work, additional, based, output] [approximate, data, exponential, log, maximizer, sample, processing, standard] [frequent, condition, small, second, top] [single, noise, sensitivity] [problem, machine, large, hypothesis, lmm, sensitive, margin, pac, function, statistical, theorem, theory, error, complexity, size, constant, convergence, fmax, class, depends, journal]
Optimal Neural Codes for Control and Estimation
Alex K. Susemihl, Ron Meir, Manfred Opper
Alex K. Susemihl, Ron Meir, Manfred Opper
[consider, solution, number, differential, set] [optimal, case, future, expected, simply, policy, depend] [based, ieee, adaptation, considering, work, coding, higher, task, dense] [process, equation, average, kalman, observation, covariance, precision, distribution, model, log, gaussian, standard, approach] [matrix, separation, noisy, principle] [control, state, sensory, tuning, encoder, time, system, observed, lqg, poisson, mmse, mutual, motor, population, neural, noise, biological, oscillator, directly, minimise, stimulus, spike, encoding] [estimation, stochastic, problem, cost, note, theory, simple, communication, function, journal, linear, framework, term, rate, distributed, constant, error, computational, depends]
Scalable Non-linear Learning with Adaptive Polynomial Expansions
Alekh Agarwal, Alina Beygelzimer, Daniel J. Hsu, John Langford, Matus J. Telgarsky
Alekh Agarwal, Alina Beygelzimer, Daniel J. Hsu, John Langford, Matus J. Telgarsky
[number, polynomial, set, degree, typically, random, high, parent, easily] [algorithm, online, adaptive, heuristic, regret, current, base, setting, bound, study, total, good, explicit] [learning, feature, datasets, training, performance, test, dataset, work, based, better, well] [figure, kernel, data, average, computationally, regression, larger, ssm, approach, standard, full] [top, running, sparse, small, ordered, fast, product] [time, weight, nonlinear, experimental, effectively, magnitude, interaction, statistically] [relative, apple, monomials, error, computational, cubic, linear, large, statistical, stochastic, method, gradient, quad, quadratic, cost, machine, monomial, shifting, implemented, lin, objective, convex, example, expansion, function, update, strong, pick, opposed, effective]
Distributed Power-law Graph Computing: Theoretical and Empirical Analysis
Cong Xie, Ling Yan, Wu-Jun Li, Zhihua Zhang
Cong Xie, Ling Yan, Wu-Jun Li, Zhihua Zhang
[dbh, replication, graph, factor, edge, degree, number, random, vertex, partitioning, workload, powergraph, hashing, shanghai, execution, existing, cutting, joseph, dmin, hash, dgc, directed, set, assignment, acm, evenly, corollary, hashed] [grid, expected, conference, sequence, max, good, result, called, strategy, bounded, social] [achieve, performance, natural, based, learning, table, china, novel] [figure, data, international, distribution, computation, university, master, model, big, parameter, parallel] [synthetic, analysis, determined, speedup, smaller] [balance, skewed, real, simultaneously, time, effectively] [method, machine, communication, distributed, theorem, function, large, theoretical, empirical, assume, note, cost]
DFacTo: Distributed Factorization of Tensors
Joon Hee Choi, S. Vishwanathan
Joon Hee Choi, S. Vishwanathan
[number, set, store, consider, existing] [algorithm, lemma, appendix, proof, version, case, conference] [datasets, table, dataset, well, reported, multiple, joint, compared, vec] [data, scaling, perform, computation, implementation, standard, scalable, model, generated] [tensor, dfacto, matrix, gigatensor, factorization, sparse, faster, requires, product, cpals, intermediate, row, cpopt, amazon, compute, denote, explosion, denotes, column, computing, cpu, vector, operator, toolbox, review, analysis, entry, extra, frontal, tamara, corresponding, yelp, alternating] [time, memory, single, behavior] [iteration, size, distributed, problem, implemented, optimization, large]
Feedback Detection for Live Predictors
Stefan Wager, Nick Chamandy, Omkar Muralidharan, Amir Najmi
Stefan Wager, Nick Chamandy, Omkar Muralidharan, Amir Najmi
[causal, random, potential, paper, increase, pilot, ratio] [feedback, detect, search, result, order, case, deployed, affect, robert, appendix, depend, intercept, goal, good, turn, concern, environment, design, study, write, methodology] [detection, natural, adding, work, idea, understanding, learn] [model, prediction, predictive, estimate, var, regression, local, amount, variance, live, data, distribution, inference, setup, relationship, average, treatment, historical, precision, allows, figure, unbiased] [add, additive, spline, small, detecting] [noise, system, time, independent, behavior, american, single] [linear, function, statistical, method, problem, suppose, journal, simple, theorem, term, predictor, estimator, form, randomized, depends, practical, theory, example]
Deep Learning for Real-Time Atari Game Play Using Offline Monte-Carlo Tree Search Planning
Xiaoxiao Guo, Satinder Singh, Honglak Lee, Richard L. Lewis, Xiaoshi Wang
Xiaoxiao Guo, Satinder Singh, Honglak Lee, Richard L. Lewis, Xiaoshi Wang
[number, paper, selection, set, procedure, random] [game, agent, uct, action, dqn, atari, policy, planning, playing, best, play, conference, current, pong, goal, combining, submarine, michigan, reward, ucttoregression, general, environment, difference] [cnn, performance, learning, training, input, deep, learned, work, convolutional, select, computer, image, layer, train, architecture, better, recognition, feature, hidden, applied, visualization, table, ale, learn, representation, ieee, idea, build, network, outperforms, human] [data, step, distribution, figure, university, approach, processing] [row] [state, neural, perception, time, three, trajectory, choice, slower, short, capture] [method, function, large, machine]
Covariance shrinkage for autocorrelated data
Daniel Bartz, Klaus-Robert Müller
Daniel Bartz, Klaus-Robert Müller
[number, high, set] [analytic, optimal, difference, chosen, benjamin] [proposed, based, training, outperforms, performance, spatial, accuracy, improve, well, tij] [sample, covariance, standard, variance, data, lag, var, figure, bias, parameter, xit, model, processing, average, larger, multivariate, highly, unbiased, imagery, dependency] [matrix, analysis, small, low, smaller, robust] [time, population, dimensionality, choice, intensity, real, motor, signal, truncation] [shrinkage, estimator, sancetta, autocorrelation, sij, xjt, large, assumption, class, estimation, increasing, consistent, cov, biased, bci, theoretical, autocorrelated, effective, squared, ledoit, sensitive, dimensional, error, theorem, constant, journal]
Orbit Regularization
Renato Negrinho, Andre Martins
Renato Negrinho, Andre Martins
[set, signed, permutahedron, region, procedure, permutation, ball, induced, concept, solution, instituto, majorization] [proposition, case, algorithm, action, decreasing, order, called, space, general, current] [matching, learning, proposed, work, ieee] [model, conditional, regression, figure, true] [seed, norm, orbitope, orbit, continuation, cone, atomic, matrix, projected, orthogonal, symmetric, sparse, vector, singular, majorized, hull, recover, orbitopes, hyperoctahedral, ngv, permutahedra, spectral, projection, absolute, denotes, minw, nuclear, analysis, role, compressed, technical, subsumes] [sorted, inverse] [group, gradient, convex, function, regularization, linear, assume, dual, regularizers, simple, problem, loss, journal, squared, arg, optimization, method, machine, framework, normal]
Do Convnets Learn Correspondence?
Jonathan L. Long, Ning Zhang, Trevor Darrell
Jonathan L. Long, Ning Zhang, Trevor Darrell
[set, consider, maximum] [grid, replaced] [keypoint, sift, convnet, feature, alignment, object, image, pose, localization, convolutional, table, trained, detection, pascal, keypoints, ground, truth, human, train, cat, stride, convnets, well, deep, conventional, correspondence, predicted, learning, bounding, pooling, visual, work, accuracy, layer, voc, better, unsupervised, computed, task, input, articulated, semantic, box, performance, intraclass, joint, despite, bird, applied, generic, location, spatial, svm, chair] [figure, prediction, perform, target, local, warped, source, prior, model, average] [compute, vector, corresponding, distance, column] [receptive, neural] [nearest, large, neighbor, size, estimation, precise]
The Noisy Power Method: A Meta Algorithm with Applications
Moritz Hardt, Eric Price
Moritz Hardt, Eric Price
[corollary, random, private, number, privacy, consider, factor, differentially, perturbation, differential, applies, apply] [algorithm, result, general, bound, case, lemma, setting, main, space, previous, probability, stronger] [work, natural, applied] [covariance, distribution, model, sample, figure, data, step, log, computation, gaussian, target] [matrix, power, singular, analysis, noisy, streaming, principal, pca, spiked, tan, orthonormal, coherence, eigenvectors, computing, top, subspace, condition, dimension, hardt, basis, requires, small, second, spectral, dominant, numerical, vector, component, completion, symmetric, roth, provided, compute, norm, alternating] [noise, time, single, represent] [method, convergence, theorem, dependence, problem, linear, theory, suppose, error, strong]
Two-Layer Feature Reduction for Sparse-Group Lasso via Decomposition of Convex Sets
Jie Wang, Jieping Ye
Jie Wang, Jieping Ye
[set, ratio, feasible, solution, existing, identify, selection, remark] [max, optimal, upper, best, case] [feature, multiple, layer, proposed, great, svm] [data, parameter, develop, regression, series, standard, estimate, step] [synthetic, sparse, exact, reduction, speedup, matrix, running, point, nonconvex, second, decomposition, vector, corresponding, determine, operator, denote] [time, society, real, rule] [sgl, tlfre, dual, screening, rejection, problem, theorem, lasso, inactive, convex, solver, group, method, solving, safe, closed, derive, duality, computational, geometric, estimation, optimization, solve, supreme, journal, note, royal, suppose, xti, regularization, machine, statistical, normal, regularizers, function, form, inf]
Predictive Entropy Search for Efficient Global Optimization of Black-box Functions
José Miguel Hernández-Lobato, Matthew W. Hoffman, Zoubin Ghahramani
José Miguel Hernández-Lobato, Matthew W. Hoffman, Zoubin Ghahramani
[entropy, median, number, random, constraint, differential, consider, easily] [search, algorithm, expected, previous, regret, optimal] [location, better, ground, learning, truth, evaluation, based, improvement, performs] [global, gaussian, bayesian, posterior, predictive, distribution, acquisition, process, maximizer, approach, figure, sample, data, approximate, sampling, latent, variance, model, hydrogen, kernel, treatment, university, expectation, multivariate, fully, hyperparameters, prior, sampled, evaluate, walker, explore, marginalize, approximated, standard] [approximation, synthetic, corresponding, compute, denote, avoid, vector] [described, noise, produced, derivative] [function, optimization, objective, cost, method, note, machine, journal, optimize, respect, domain, arg, active, example]
Submodular meets Structured: Finding Diverse Subsets in Exponentially-Large Structured Item Sets
Adarsh Prasad, Stefanie Jegelka, Dhruv Batra
Adarsh Prasad, Stefanie Jegelka, Dhruv Batra
[label, set, hamming, marginal, item, factor, graph, map, ball, number, scoring, list, cardinality, divmbest, labelings, high, dlb, graphical, augmentation, random, solution, maximization, covered, labeling, quality, user, edge, node] [submodular, gain, algorithm, base, monotone, relevance, oracle, maximizing, space, exponentially, lemma, general, bound, best, simply, maximize, previous] [diversity, structured, segmentation, diverse, image, presence, ground, table, work, pascal, pair, output, object, natural, report, combined, learning] [inference, step, model, approximate, prediction] [greedy, approximation, nonnegative, provided, small] [single, hop, three, rare] [function, group, large, term, machine, interactive, cost, problem]
A Differential Equation for Modeling Nesterov’s Accelerated Gradient Method: Theory and Insights
Weijie Su, Stephen Boyd, Emmanuel Candes
Weijie Su, Stephen Boyd, Emmanuel Candes
[solution, random, set, discrete, energy, continuous, differential, unique, consider, composite, mathematical, number] [proof, initial, bound, study, replaced, general, starting, difference] [proposed, original, unit, performance] [scheme, step, parameter, gaussian, equation, global, standard] [numerical, matrix, condition, small, siam, second, analysis, sparse, equivalent, suggests, singular, denotes] [speed, time, derivative, stable, introduction, velocity] [ode, theorem, gradient, convergence, restarting, convex, rate, method, linear, size, accelerated, inequality, obeys, theory, objective, restarted, optimization, functional, function, constant, tsr, generalized, journal, positive, lipschitz, srn, ordinary, min, kmin, proximal, grn, smooth, equivalence, large, strong, family, convexity, princeton, note, empirically]
Learning Distributed Representations for Structured Output Prediction
Vivek Srikumar, Christopher D. Manning
Vivek Srikumar, Christopher D. Manning
[label, structure, set, discrete, random, list] [associated, sequence, algorithm, order, conference, goal] [feature, learning, structured, input, compositional, istro, output, multiclass, training, representation, performance, labeled, natural, language, vec, tagging, baseline, table, purely, basque, newsgroup, propose, english, svm, compositionality, jointly, represented, task, separate, work] [model, prediction, standard, figure, inference, data, equation, allow, markov] [vector, tensor, atomic, matrix, alternating, rank, minimization, product, denote, norm] [weight, shared, dimensionality, semantically, represent, document, share] [function, objective, problem, structural, distributed, dimensional, machine, loss, method, statistical]
On Sparse Gaussian Chain Graph Models
Calvin McCarter, Seyoung Kim
Calvin McCarter, Seyoung Kim
[graph, structure, graphical, set, undirected, moralization, directed, genomic, mapping, correspond, performing, random, indirect, procedure] [conference, goal, order, alternative, considered, previous] [multiple, structured, learning, test, err, supervised, work, performance, input, output, joint, pattern, dataset, jointly] [chain, model, gaussian, data, figure, cggm, regression, markov, cggms, prediction, conditional, approach, inference, simulation, probabilistic, precision, gene, moralized, distribution, lwf, true, marginalization, byz, estimate, international, covariance, dotted, bxy, multivariate] [sparse, component, sparsity, recovery] [recall, estimated, inferred] [linear, estimation, functional, problem, optimization, journal]
On the Information Theoretic Limits of Learning Ising Models
Rashish Tandon, Karthikeyan Shanmugam, Pradeep K. Ravikumar, Alexandros G. Dimakis
Rashish Tandon, Karthikeyan Shanmugam, Pradeep K. Ravikumar, Alexandros G. Dimakis
[graph, set, number, pmax, consider, random, edge, subset, corollary, pavg, ising, provide, degree, graphical, tanh, approx, covering, node, entropy, remark, differ, high, connected, girth, undirected, requirement] [bound, lower, lemma, probability, bounded, max, difference, deterministic, case, proof, exponentially, general, upper, uniform, simply] [pair, learning, ieee] [log, sample, distribution, conditional, markov, allows, approach, exponential, scaling, model] [small, smaller, corresponding, low, denote, close, requires] [correlation, mutual, length, underlying, scalar] [class, structural, large, note, complexity, key, measure, theorem, term, error, statistical, problem, dependence, estimator, constant, family, derive, estimation, simple]
Learning with Pseudo-Ensembles
Phil Bachman, Ouais Alsharif, Doina Precup
Phil Bachman, Ouais Alsharif, Doina Precup
[parent, penalty, random, binary, tree, notion, set, existing] [result, case, minimize] [learning, dropout, pea, training, performance, output, network, labeled, layer, sentiment, feature, rntn, masking, ensemble, hidden, mnist, deep, well, work, trained, convolutional, phrase, softmax, fuzzing, table, supervised, manifold, indicates, recursive, applied, adding, original, denoising, spawned] [model, standard, data, distribution, source, child, gaussian, unlabeled, full, process, sampling, generated, target, sampled, regression, tangent] [robust, vector, subspace, randomly, tensor, analysis, improved, matrix] [noise, neural, weight, measured] [regularization, regularizer, linear, loss, optimization, averaged, method, size, objective, stochastic]
An Autoencoder Approach to Learning Bilingual Word Representations
Sarath Chandar A P, Stanislas Lauly, Hugo Larochelle, Mitesh Khapra, Balaraman Ravindran, Vikas C. Raykar, Amrita Saha
Sarath Chandar A P, Stanislas Lauly, Hugo Larochelle, Mitesh Khapra, Balaraman Ravindran, Vikas C. Raykar, Amrita Saha
[binary, set, sum, procedure, number] [conference, annual, considered, context, simply] [language, word, training, bilingual, learning, reconstruction, representation, learn, klementiev, sentence, embeddings, trained, autoencoder, work, train, propose, association, natural, translation, aligned, multilingual, meeting, monolingual, network, labeled, extracted, performance, proposed, input, rely, original, learned, vectorial, reconstructing, text, joint, english, test, hermann, linguistics, additional, annotated, pair, based, christopher, report, table] [data, model, approach, processing, international, parallel, figure] [vector, reconstruct] [document, neural, decoder, encoder, described, meaningful, correlation, investigate] [computational, machine, size, method, stochastic, term, loss, regularization]
Deep Convolutional Neural Network for Image Deconvolution
Li Xu, Jimmy S. Ren, Ce Liu, Jiaya Jia
Li Xu, Jimmy S. Ren, Ce Liu, Jiaya Jia
[structure, random, existing, number, partially, mapping, energy, consider] [previous, result] [deconvolution, image, network, cnn, convolutional, denoise, convolution, blur, training, learning, deep, input, architecture, spatial, visual, natural, based, trained, camera, degradation, well, train, saturated, hidden, blurry, compression, psnrs, disk, denoising, work, motion, performance, traditional, handle, blurred, decent, stacked, layer, higher, supervised, restoration, understanding, cho, challenging] [kernel, separable, model, figure, generative, inversion, wiener, approach, pseudo, gaussian, generate, expressive] [initialization, sparse, small, sharp, singular, fourier] [neural, inverse, noise, system, nonlinear, directly, expressed] [method, size, large, note, strong, function, journal, address]
Do Deep Nets Really Need to be Deep?
Jimmy Ba, Rich Caruana
Jimmy Ba, Rich Caruana
[number, set, increase, connected, paper] [conference, gap, probability] [deep, shallow, mimic, trained, accuracy, training, teacher, train, net, learning, accurate, convolutional, hidden, original, timit, layer, ensemble, learn, compression, pooling, dropout, cnn, learned, complex, input, convolution, better, recognition, logits, test, dnn, deeper, ecnn, cnns, performance, architecture, network, table, continues, work, labeled, help, depth, achieve, ieee, improve, tiny, representational, output, dev, feature, predicted] [model, data, unlabeled, perform, fully, figure, international, standard] [student, matrix, small, suggests] [neural, speech, weight, single, three] [large, linear, function, regularization, size, loss, machine, stochastic, cost]
Sampling for Inference in Probabilistic Models with Fast Bayesian Quadrature
Tom Gunter, Michael A. Osborne, Roman Garnett, Philipp Hennig, Stephen J. Roberts
Tom Gunter, Michael A. Osborne, Roman Garnett, Philipp Hennig, Stephen J. Roberts
[entropy, interval, consider, set, existing, graph, marginal, typically, random, binary] [expected, space, utility, selecting, optimal, probability] [ground, truth, learning, input, performed, output, transform, compared, evaluation] [likelihood, wsabi, model, sampling, bayesian, prior, posterior, figure, variance, quadrature, monte, sample, smc, carlo, gaussian, integral, integrand, true, covariance, inference, approach, uncertainty, linearisation, computation, process, probabilistic, average, bmc, moment, mass, arrive, regression, university, standard, estimate, generate, log, scheme, cheap, mixture, predictive] [synthetic, approximation, uniformly, reduction, corresponding] [time, directly, conditioned, range, real, dynamic, drawn, described] [function, active, error, machine, convergence, cost, simple, relative, problem]
Testing Unfaithful Gaussian Graphical Models
De Wen Soh, Sekhar C. Tatikonda
De Wen Soh, Sekhar C. Tatikonda
[graphical, node, set, graph, relation, faithful, unfaithful, connected, path, faithfulness, linearly, exists, edge, partition, random, imply, consider, provide, topology, variable, undirected, paper, subset, number] [algorithm, implies, proof, lemma, main, case, result, observe, associated, search, study] [test, output, pair, patch, testing, natural] [conditional, independence, gaussian, distribution, covariance, concentration, markov, global, local, multivariate, step, model, precision, figure, university, series] [matrix, vector, column, submatrix, denote, running, condition, row, separation] [distinct, independent, dependent, time] [separator, example, linear, theorem, suppose, form, note, assume, positive, including, size, assumption, class]
Median Selection Subset Aggregation for Parallel Inference
Xiangyu Wang, Peichao Peng, David B. Dunson
Xiangyu Wang, Peichao Peng, David B. Dunson
[median, message, selection, fullset, bolasso, subset, number, set, inclusion, partition, prob, gic, high, variable, excellent, extensive, excluded] [algorithm, probability, case, combining, aggregation, chosen, exp, selecting] [feature, performance, select, selected, learning, accuracy, well, produce, training, multiple] [model, data, sample, average, parallel, full, true, prediction, figure, averaging, regression, inference, bootstrap, log, approach, david, volume, computation] [square, vector, arxiv, preprint, low, sparse] [time, estimated, implement] [size, lasso, computational, error, linear, consistency, statistical, distributed, machine, communication, rate, theoretical, journal, logistic, estimation, simple, optimization, usual, theorem, assume, large, consistent, method]
Pre-training of Recurrent Neural Networks via Linear Autoencoders
Luca Pasa, Alessandro Sperduti
Luca Pasa, Alessandro Sperduti
[solution, number, set, procedure] [sequence, algorithm, optimal, good, space, case, initial, start, return] [training, rnn, hidden, autoencoder, learning, accuracy, test, input, performance, proposed, dataset, whidden, deep, nottingham, muse, network, jsb, based, datasets, validation, reported, autoencoders, trained] [data, computation, prediction, approach, approximate, allows, full, covariance, model] [matrix, principal, svd, decomposition, rank, component, exact, compute, singular, analysis, approximation, pret, rnd, woutput] [neural, time, state, recurrent, dynamical, described, polyphonic] [linear, epoch, function, problem, large, size, gradient]
Weighted importance sampling for off-policy learning with linear function approximation
A. Rupam Mahmood, Hado P. van Hasselt, Richard S. Sutton
A. Rupam Mahmood, Hado P. van Hasselt, Richard S. Sutton
[solution, discrepancy, consider, differ, ratio, high] [policy, case, reinforcement, general, algorithm, difference, terminal, conference, provisional, environment, incrementally, action, agent, probability, tabular] [learning, feature, based, conventional, task, performance, joint] [estimate, sampling, target, distribution, parameter, monte, carlo, mse, sample, conditional, variance, fully, expectation, international, shift, covariate] [approximation, equivalent, vector, second, estimating, matrix] [time, state, behavior, weight, individual] [function, empirical, method, error, weighted, linear, theorem, sutton, ois, weighting, machine, estimator, form, objective, journal, computational, large, consistent, corrected, squared, eligibility, regularization, min]
Discovering Structure in High-Dimensional Data Through Correlation Explanation
Greg Ver Steeg, Aram Galstyan
Greg Ver Steeg, Aram Galstyan
[corex, structure, number, random, tree, set, personality, variable, solution, uncorrelated, consider, dna, hierarchy] [max, probability, appears, total, bound] [learning, representation, learned, hidden, labeled, learn, newsgroup, perfectly, text, human, diverse, accuracy, unsupervised, yoshua, complex, automatically, test] [data, latent, multivariate, model, approach, true, bottleneck, figure, standard, distribution, generated] [recover, clustering, synthetic, second, small, comp, survey, exactly] [mutual, hierarchical, correlation, neural, perfect, common, described, explanation, ica, sci, explain, three, independent, broad, twenty, single] [optimization, objective, method, note, measure, problem, theoretical, machine, linear, optimizing, complexity, group, written, journal]
Model-based Reinforcement Learning and the Eluder Dimension
Ian Osband, Benjamin Van Roy
Ian Osband, Benjamin Van Roy
[set, provide, number, high, random, consider, corollary, programming, existing, discrete, sum, apply] [regret, reinforcement, mdp, eluder, algorithm, optimal, dime, write, policy, bound, psrl, proposition, cient, reward, general, wit, transition, future, lemma, benjamin, expected, van, mdps, action, space, sequence, michael, appendix, agent, probability, deviation, follow, satisfy, bellman, cumulative, extend, optimistic, bounded, daniel, wftk] [learning, natural, art] [distribution, posterior, sampling, global, markov, sequential, prior, equation] [dimension, analysis, notation, condition] [state, time, control, underlying, dynamic] [function, machine, linear, lipschitz, problem, theorem, require, complexity, constant, journal, quadratic, large, optimize]
Online Decision-Making in General Combinatorial Spaces
Arun Rajkumar, Shivani Agarwal
Arun Rajkumar, Shivani Agarwal
[polytope, set, consider, permutation, entropy, mapping, spanning, birkhoff] [online, algorithm, decision, combinatorial, regret, ldomd, general, submodular, bound, unnormalized, polytopes, transportation, case, matroid, omd, setting, space, receives, selects, associated, base, hedge, study, extreme, appendix, demand, supply, goal, considered, legendre, minimize, studied, incurred] [learning, performed, representation, well, represented, natural] [step, figure, distribution] [vector, decomposition, projection, suitable, boolean, matrix, point, special, component] [time, trial, independent] [loss, convex, function, linear, example, negative, relative, problem, descent, journal, note, theorem, mirror, generalizes, applying, cost, framework, bregman, closed, optimization]
Sparse Multi-Task Reinforcement Learning
Daniele Calandriello, Alessandro Lazaric, Marcello Restelli
Daniele Calandriello, Alessandro Lazaric, Marcello Restelli
[set, number, exists, consider] [reinforcement, space, optimal, fqi, policy, algorithm, bounded, mdps, action, max, reward, setting, transition, study, mdp, lemma, exploit, previous, fwk, total, history, bellman, alessandro, maxa, player, dealer] [learning, performance, feature, representation, level, transformation, jointly, multiple, hand, task, similarity, transfer, learn, joint, improve] [target, regression, distribution, sample, full, perform, approximate] [sparsity, matrix, sparse, vector, small, approximation, rank, corresponding, introduce, low, operator] [weight, shared, state, expect] [assumption, function, iteration, linear, min, problem, lasso, assume, large, solving, arg, loss, machine, objective, optimization, solve, theorem]
Coresets for k-Segmentation of Streaming Data
Guy Rosman, Mikhail Volkov, Dan Feldman, John W. Fisher III, Daniela Rus
Guy Rosman, Mikhail Volkov, Dan Feldman, John W. Fisher III, Daniela Rus
[coreset, set, number, coresets, construction, bitcoin, paper, quality, vassar, sum, usa, random, balanced] [algorithm, price, space, main, result, uniform, optimal] [video, segmentation, input, summarization, computed, well, segment, feature, proposed, stream, original, work, representation, mit, image, based, output, compression] [data, figure, gps, demonstrate, supplementary, approximate, parallel, sample, allows, approximates, approach, approximated, local, log] [running, streaming, approximation, point, compute, analysis, small, dimension, computing, synthetic, provable, reduction, compact, reckoning] [time, single, signal, dynamic, content, dimensionality, desired, meaningful, represents] [cost, linear, large, error, size, problem, theorem, function, squared, example, cubic]
Learning the Learning Rate for Prediction with Expert Advice
Wouter M. Koolen, Tim van Erven, Peter Grünwald
Wouter M. Koolen, Tim van Erven, Peter Grünwald
[factor, number] [llr, regret, cumulative, grid, adahedge, bound, algorithm, lemma, gap, expert, bounded, round, minimum, budget, guarantee, sequence, hedge, best, strategy, imax, appears, optimal, chosen, flipflop, lower, choosing, conservative, learner, appendix, ftl, probability, aah, play, case, played, previous, setting] [learning, higher, better, performs, well, additional, work, learn] [exponential, data, prior, distribution, prediction, figure, approach, standard, mix] [small, denote, introduce, worse, smaller, approximation, special, point] [time, choice, tuning, range, single] [rate, mixability, loss, active, large, addition, essentially, journal, constant]
Subspace Embeddings for the Polynomial Kernel
Haim Avron, Huy Nguyen, David Woodruff
Haim Avron, Huy Nguyen, David Woodruff
[polynomial, mapping, number, random, high, set, hash, induced, degree] [space, lemma, algorithm, probability, oblivious, version, conference, design, goal, explicit] [learning, feature, embedding, propose, applied, transform, entire, training, extracted, explicitly, based, embeddings, svm, table, representation, better] [kernel, approximate, regression, data, international] [matrix, ketch, ensor, subspace, principal, pca, compute, component, computing, approximation, product, rank, sketching, low, singular, orthonormal, ount, analysis, pagh, fast, faster, vector, top, column, ose, row, pham, material, close] [time, independent, simultaneously] [linear, theorem, form, problem, statistical, note, generalization, error, solve, function, sign, method, regularization]
Efficient Minimax Strategies for Square Loss Games
Wouter M. Koolen, Alan Malek, Peter L. Bartlett
Wouter M. Koolen, Alan Malek, Peter L. Bartlett
[ball, consider, constraint, maximum, recurrence, apply, induction, feasible] [minimax, game, strategy, maximin, action, mahalanobis, lemma, regret, brier, vmax, case, learner, max, outcome, online, best, optimal, play, proof, backwards, conference, bound, swap, servedio, berkeley, hindsight, chooses, minimizer, base, probability, aek] [learning, unit, work] [prediction, allows, university, figure, exponential, sequential, data, gaussian, maximizer, likelihood, computationally, distribution] [symmetric, second, analysis, simplex, matrix, condition, euclidean, largest, vector, distance, eigenvalue, norm, square, optimizer, point] [state, time, length] [loss, inf, squared, quadratic, theorem, function, objective, optimization, peter, derive, min, problem, normalized, calculate, linear, closed]
Delay-Tolerant Algorithms for Asynchronous Distributed Online Learning
Brendan McMahan, Matthew Streeter
Brendan McMahan, Matthew Streeter
[set, sum, consider, john, feasible] [algorithm, bound, regret, online, adaptive, lemma, general, appendix, order, search, optimal, choose, difference, main] [learning, performance, applied, dataset, accuracy, scale, work, better, training, based] [data, standard, model, figure, prediction, parameter, parallel, scaling, computation, plot, minibatch] [analysis, exactly, read, analyze, sparse, vector, second] [time, reader, optimized] [update, gradient, delay, adaptiverevision, rate, loss, fwd, coordinate, gfwd, bck, stochastic, large, hypback, descent, gbck, distributed, convex, function, hypothetical, inorder, achieves, machine, constant, martin, adagrad, linear, assumption, key, asyncadagrad, prove, example, practice, outstanding, note, updaters, duchi]
A Probabilistic Framework for Multimodal Retrieval using Integrative Indian Buffet Process
Bahadir Ozdemir, Larry S. Davis
Bahadir Ozdemir, Larry S. Davis
[hashing, binary, set, number, user, procedure, high] [relevance, feedback, conference, customer, space, probability, search, ranking] [image, retrieval, visual, dataset, abstract, feature, multimodal, textual, computer, semantic, similarity, ieee, learning, integrative, based, pattern, proposed, sun, recognition, vision, june, znk, revised, retrieved, performance, computed, ground, truth, text, compared, indicating, category, training, object] [data, model, latent, approach, figure, prior, processing, distribution, precision, indian, international, ibp, sample, likelihood, expectation, sampling, inference, buffet, probabilistic, dish, process] [matrix, vector, top, spectral] [recall, neural, system] [query, method, machine, framework, large, class]
General Stochastic Networks for Classification
Matthias Zöhrer, Franz Pernkopf
Matthias Zöhrer, Franz Pernkopf
[number, experiment, permutation] [general, best, case, conference, static, criterion, order, includes] [learning, training, mnist, network, walkback, test, deep, gsn, layer, labeled, supervised, hybrid, discriminative, input, multiple, dataset, additional, boltzmann, applied, gsns, achieved, dropout, table, rotated, odd, trained, convolutional, hidden, outperform, baseline, background, reconstruction, representation, performance, invariant, output, datasets] [generative, figure, model, processing, markov, annealed, parameter, restricted, gaussian, international, chain, setup] [introduce, denotes, second, introduced] [neural, noise, dynamic, three, speech, time] [stochastic, machine, function, optimization, cost, error, loss, variant, generalization, objective, update, regularization]
Provable Submodular Minimization using Wolfe's Algorithm
Deeparnab Chakrabarty, Prateek Jain, Pravesh Kothari
Deeparnab Chakrabarty, Prateek Jain, Pravesh Kothari
[set, polytope, polynomial, number, solution, procedure, subset, arbitrary, vertex, consider, graph, apply, claim] [algorithm, submodular, major, minor, minimizer, good, minimum, base, sfm, lemma, cycle, satoru, implies, fujishige, minimizing, combinatorial, version, proof, wolfe, iwata, ellipsoid, stefanie, jeff, bound, start, optimal, rishabh, renumber, tth, best, associated, oracle] [work, submodularity, learning] [figure, approximate] [point, norm, analysis, running, minimization, robust, fast, vector, greedy, exactly] [time, described] [theorem, function, prove, iteration, method, linear, note, assume, arg, simple, optimization, dependence, convex, update, large, proving, satisfying, empirical, suppose, practical, solves]
Efficient Partial Monitoring with Prior Information
Hastagiri P. Vanchinathan, Gábor Bartók, Andreas Krause
Hastagiri P. Vanchinathan, Gábor Bartók, Andreas Krause
[partial, set, feasible, adversarial, consider, existing] [action, outcome, regret, opponent, learner, algorithm, feedback, game, monitoring, chooses, confidence, expected, minimax, ellipsoid, definition, observable, bpm, pricing, best, bound, chosen, probability, price, case, difference, choose, version, fixed, receives, optimality, cumulative, finite, easy, feedexp, suffered, upper, bor, iid, goal, optimal] [performance, based, learning, unit] [distribution, prior, gaussian, true, track, step, local, figure, posterior, covariance, sample] [matrix, cbp, condition, product] [time, cell, dynamic, real, tracking, signal, rule] [loss, stochastic, problem, locally, update, assume, linear, note, constant, achieves, family, large, rate, method, example]
On Prior Distributions and Approximate Inference for Structured Variables
Oluwasanmi O. Koyejo, Rajiv Khanna, Joydeep Ghosh, Russell Poldrack
Oluwasanmi O. Koyejo, Rajiv Khanna, Joydeep Ghosh, Russell Poldrack
[set, selection, corollary, consider, solution, entropy, high, subset, constraint, maximum, random] [base, restriction, case, probability, optimal, submodular, space, design] [structured, performance, selected, proposed, test, propose, training] [density, prior, inference, approach, data, posterior, distribution, approximate, bayesian, slab, predictive, regression, ard, log, gaussian, spikeslabkl, standard, model, tested, voxels, forward, generated, sample, restricted, fmri, variational] [projection, support, sparse, neuroimaging, low, condition, recovery, greedy, ridge, special, denotes, rank, sparsity, approximation, product] [spike, brain, snr, noise, auc, represent] [domain, min, theorem, function, lasso, arg, convex, functional, measure, relative, dimensional, normalized, framework, structural, denoted, respect, size, assumption]
Identifying and attacking the saddle point problem in high-dimensional non-convex optimization
Yann N. Dauphin, Razvan Pascanu, Caglar Gulcehre, Kyunghyun Cho, Surya Ganguli, Yoshua Bengio
Yann N. Dauphin, Razvan Pascanu, Caglar Gulcehre, Kyunghyun Cho, Surya Ganguli, Yoshua Bengio
[high, random, region, constraint, solution, number] [algorithm, order, exponentially, search, minimum, conference, result, qualitatively] [learning, deep, network, training, natural, hidden, mnist, trained, proposed] [local, curvature, step, gaussian, approach, distribution, model, global, larger, international] [point, hessian, matrix, small, eigenvalue, approximation, distance, review, taylor, exact, second, subspace, krylov, dean, absolute] [neural, recurrent, inverse, dimensionality, single, classical] [saddle, error, newton, critical, method, gradient, negative, descent, optimization, trust, function, dimensional, sgd, problem, theory, large, direction, generalized, statistical, rapidly, escape, rescaling, bray, expansion, positive, machine, distributed, plateau, linear, increasing]
Generalized Unsupervised Manifold Alignment
Zhen Cui, Hong Chang, Shiguang Shan, Xilin Chen
Zhen Cui, Hong Chang, Shiguang Shan, Xilin Chen
[integer, structure, set, solution, path, geodesic, discover] [algorithm, space, relax, explicit, best] [manifold, alignment, matching, guma, unsupervised, face, embedding, learning, geometry, image, itl, adaptation, orifea, feature, proposed, datasets, recognition, performance, visual, learn, accuracy, align, well, original, build, propose, based, considering, cross, dataset, neighborhood, aligned, wang, better, effectiveness, achieve, pose] [local, data, target, source, global, sample, model, kernel] [matrix, corresponding, compare, preserving, distance, protein, compact, ith, adjacency, point] [mutual, weight, intrinsic] [domain, method, objective, optimization, function, problem, convex, term, min, generalized, quadratic, cost]
Unsupervised Transcription of Piano Music
Taylor Berg-Kirkpatrick, Jacob Andreas, Dan Klein
Taylor Berg-Kirkpatrick, Jacob Andreas, Dan Klein
[discrete, symbolic, structure, collection] [play, best, total, gain, order] [activation, unsupervised, ieee, discriminative, segment, baseline, task, test, hidden, joint, score] [model, distribution, figure, generated, parameter, inference, generate, process, international, approach, standard, estimate, full, conditional, step, markov, prior] [spectral, component, matrix, vector, denote] [piano, musical, time, transcription, spectrogram, music, event, envelope, system, audio, onset, frequency, signal, noise, polyphonic, song, parameterized, midi, velocity, instrument, acoustic, conditioned, duration, poisson, produced, timbral, observed, single, freq, sound, benetos, depicted, length, corpus, speech, generates, rise, published, choice] [note, string, achieves, key, update, gradient, positive, example]
Combinatorial Pure Exploration of Multi-Armed Bandits
Shouyuan Chen, Tian Lin, Irwin King, Michael R. Lyu, Wei Chen
Shouyuan Chen, Tian Lin, Irwin King, Michael R. Lyu, Wei Chen
[set, collection, maximization, number, spanning, identifying, correspond, radius, provide, matchings] [decision, arm, algorithm, combinatorial, bound, exploration, clucb, exchange, lower, pure, optimal, best, csar, cpe, reward, upper, budget, bandit, general, expected, setting, learner, probability, radt, called, logarithmic, game, round, bubeck, oracle, maxm, gap, matroids, phase, result, encompasses, online, matroid, access, mabs, call, type] [learning, novel, work, challenge] [sample, distribution, supplementary, gaussian] [analysis, vector, denote, largest, material] [independent, formally] [complexity, class, problem, stochastic, including, framework, theorem, empirical, objective, arg, constrained, assume, example, constant, error, applying, conjecture, suppose]
Non-convex Robust PCA
Praneeth Netrapalli, Niranjan U N, Sujay Sanghavi, Animashree Anandkumar, Prateek Jain
Praneeth Netrapalli, Niranjan U N, Sujay Sanghavi, Animashree Anandkumar, Prateek Jain
[set, hard, number, consider, perturbation, random, tight, remove] [algorithm, deterministic, best, proof, goal, result] [input, level, background, reconstruction, work, achieve, better, frame, additional] [figure, step, parameter, global, standard] [rank, matrix, sparse, robust, pca, low, ialm, altproj, sparsity, alternating, recovery, thresholding, singular, projection, running, incoherence, stage, supp, required, synthetic, exact, approximation, separation, involves, decomposition, sujay, norm, smaller, component, establish, match, faster, consists, requires, top, eigenvectors, escalator, symmetric] [time, fraction, varying] [method, convex, error, problem, convergence, note, complexity, large, theorem, require, threshold, size, assumption, microsoft]
Spectral Methods meet EM: A Provably Optimal Algorithm for Crowdsourcing
Yuchen Zhang, Xi Chen, Dengyong Zhou, Michael I. Jordan
Yuchen Zhang, Xi Chen, Dengyong Zhou, Michael I. Jordan
[number, worker, crowdsourcing, labeling, label, zaj, binary, random, item, set, web, provide] [algorithm, optimal, probability, lower, proposition, minimax, majority, bounded, voting, initial, guarantee] [learning, proposed, work, propose, performance, datasets, table] [estimate, true, model, approach, likelihood, step, latent] [confusion, matrix, spectral, stage, zbj, compute, tensor, estimating, initialization, dawid, second, column, assumes, compare, trec, arxiv, entry, diagonal, skene, qjl, wmin, requires, small, thresholding] [zij, three, scalar, observed, real, choice] [iteration, theorem, estimator, method, error, min, convergence, theoretical, assumption, empirical, assume, rate, function, class, problem, constant, positive]
Exploiting easy data in online optimization
Amir Sani, Gergely Neu, Alessandro Lazaric
Amir Sani, Gergely Neu, Alessandro Lazaric
[edge, set, consider, corollary, assignment, number, structure] [regret, algorithm, rod, online, decision, setting, ftl, lip, best, expert, bound, strategy, benchmark, learner, worst, guarantee, lop, general, total, open, dge, easy, suffers, rooij, result, suggested, bandit, case, difference, version, tuned, observe, cumulative, environment, round, upper, order, probability, hedging, warmuth, ucb, conference, expected, sequence, appropriately, history] [learning, performance, proposed, achieve, additional, well, work] [log, prediction, parameter, figure, distribution] [improved, analysis] [time, simultaneously, directly, tracking] [loss, problem, convex, optimization, theorem, machine, constant, variant, achieves, complexity, note, assume, theoretical]
Learning Mixtures of Ranking Models
Pranjal Awasthi, Avrim Blum, Or Sheffet, Aravindan Vijayaraghavan
Pranjal Awasthi, Avrim Blum, Or Sheffet, Aravindan Vijayaraghavan
[permutation, polynomial, set, random, partition, consider, number] [algorithm, ranking, probability, position, central, case, ranked, access, lemma, ecomp, order, subsample, main, interesting, ecover] [learning, learn, work, based, representative, appearing, handle] [mixture, model, moment, estimate, supplementary, sampled, generated, full, degenerate, allows, data, university, true, pijk, parameter, popular, distribution] [tensor, element, rank, denote, vector, second, distance, pairwise, ensor, top, recover, decomposition, spectral, separation, material, retrieve, matrix, analysis, refer, wmin] [single, distinct, time, reader, system, underlying] [problem, linear, theory, machine, key]
From Stochastic Mixability to Fast Rates
Nishant A. Mehta, Robert C. Williamson
Nishant A. Mehta, Robert C. Williamson
[random, variable, set, high, exists, unique, consider, apply, paper, notion] [probability, result, proof, case, implies, general, upper, space, bound, optimal, lemma, van, bounded, oracle, appendix, sense, open, conference, minimizer, robert, online, best, extend] [learning, work, select, conv] [log, moment, perspective, australian, target, prediction, approach] [condition, fast, analysis, minimization, exact, special, nonnegative] [direct, slow, control] [mixability, stochastic, theorem, statistical, risk, erm, function, bernstein, constant, loss, excess, problem, theory, empirical, class, rate, annals, minimizers, convex, peter, stochastically, convexity, inequality, measure, note, mixable, mendelson, large, shahar, erven, characterizes, assumption]
Beyond Disagreement-Based Agnostic Active Learning
Chicheng Zhang, Kamalika Chaudhuri
Chicheng Zhang, Kamalika Chaudhuri
[label, set, number, candidate, provide, applies, exists, binary, region, assigned] [algorithm, oracle, bound, probability, case, guarantee, space, lemma, coverage, version, goal, general, total, main, setting, observe, uniform, optimal, minimizer, adaptive, expected] [learning, better, achieve, work, selective, performance, based, novel, predicted] [target, distribution, data, true, prediction, sample, sampling] [denote, analysis, dimension, provided] [noise] [active, predictor, error, hypothesis, agnostic, complexity, excess, realizable, example, class, disagreement, query, guaranteed, labelling, risk, unlabelled, respect, linear, suppose, labelled, errd, generalization, queried, consistent, key, epoch, abstention, theorem, problem, achieves, passive, function, constant, generally]
Optimal Regret Minimization in Posted-Price Auctions with Strategic Buyers
Mehryar Mohri, Andres Munoz
Mehryar Mohri, Andres Munoz
[set, number, node, consider, fact, factor, path, tree, left, exists, pfs, binary, notion] [algorithm, regret, price, buyer, seller, monotone, strategic, offered, bound, upper, revenue, valuation, lower, proposition, surplus, bandit, truthful, admits, rejected, offer, design, favorable, pfsr, scenario, reserve, amin, best, previous, bounded, sublinear, search, optimal, game, facing, sequence, repeated, round, logarithmic, study, accept, decreasing, order, combining, increment, online, phase, lemma, mon, strategy, advertisement, discounted, accepted] [learning, achieve, natural, view] [log, parameter, figure] [analysis, analyze, knowledge, introduced, compare, minimization] [presented, time, choice, behavior] [theorem, family, optimization, assumption, problem, journal, consistent, inequality, convex]
Rates of Convergence for Nearest Neighbor Classification
Kamalika Chaudhuri, Sanjoy Dasgupta
Kamalika Chaudhuri, Sanjoy Dasgupta
[set, notion, marginal, radius, consider, ball, random, cover] [probability, bound, space, general, lower, decision, upper, bounded, ern, best, case, starting, studied, expected, universal, study, open, lie] [work, training, ieee, understanding, based] [distribution, data, nonparametric, mass, conditional, prediction, figure, standard, expectation] [condition, point, denote, support] [choice, earlier, centered, behavior, range, science, common] [nearest, neighbor, convergence, metric, smoothness, rate, boundary, constant, function, risk, effective, measure, consistency, theorem, class, respect, error, margin, annals, assumption, holder, smooth, pick, positive, generalization, prx, suppose, excess, query, statistical, measurable, usual]
Fast Prediction for Large-Scale Kernel Machines
Cho-Jui Hsieh, Si Si, Inderjit S. Dhillon
Cho-Jui Hsieh, Si Si, Inderjit S. Dhillon
[solution, number, reduce, polynomial, set, approximating, consider, random, apply] [algorithm, decision, appendix, optimal, general, upper, bound, focus, design] [svm, training, proposed, based, accuracy, computed, select, testing, learning, original, better, test, table, applied, learn] [kernel, prediction, approximate, pseudo, approach, regression, figure, model, local, sample, data] [landmark, approximation, fast, kmeans, ridge, matrix, compute, vector, support, ntest, ntrain, comparison, faster, point, requires, smaller, compare, computing, basis, clustering] [time, reduced, experimental, speed, cluster, presented] [error, weighted, method, linear, ldkl, function, problem, note, cost, achieves, large, framework, solver, form, solved]
Latent Support Measure Machines for Bag-of-Words Data Classification
Yuya Yoshikawa, Tomoharu Iwata, Hiroshi Sawada
Yuya Yoshikawa, Tomoharu Iwata, Hiroshi Sawada
[set, consider, high, number] [space, current, result] [smm, word, learning, bow, representation, represented, performance, accuracy, smms, training, appearing, proposed, svms, discriminative, embeddings, svm, webkb, text, embedding, medlda, based, visualization, separating, visualizing, achieve, indicates, effectiveness, learnt, work] [latent, kernel, distribution, figure, parameter, data, gaussian, standard, estimate] [support, vector, ith, low, rank, robust, hilbert] [document, topic, dimensionality, vocabulary, represent, varying, newsgroups, neural, semantically] [method, measure, class, problem, machine, rbf, occurrence, gradient, dimensional, respect, characteristic, linear, empirical, margin, journal, solve, large]
Message Passing Inference for Large Scale Graphical Models with High Order Potentials
Jian Zhang, Alex Schwing, Raquel Urtasun
Jian Zhang, Alex Schwing, Raquel Urtasun
[layout, message, passing, region, graphical, dcbp, set, graph, agreement, claim, energy, connected, random, factor, program, belief, sum, scoring, partition, assigned, map, newly, densely, marginals, superpixel, entropy, high] [algorithm, order, case, exchange, observe, max] [task, segmentation, additional, joint, computer, despite, proposed, semantic, score, original, scene, multiple] [inference, approach, local, computation, figure, model, marginalization, converge, global, parallel, variational] [cbp, decomposition, introduce, required, second, largest, pairwise, corresponding, faster, estimating] [single, magnitude] [dual, distributed, primal, convex, function, constant, computational, size, note, large, complexity, linear, convergence, duality, normalized]
Automatic Discovery of Cognitive Skills to Improve the Prediction of Student Learning
Robert V. Lindsey, Mohammad Khajah, Michael C. Mozer
Robert V. Lindsey, Mohammad Khajah, Michael C. Mozer
[labeling, number, set, discovery, assigned] [expert, probability, customer, sequence, uniform, crp, expected, transition] [table, learning, performance, datasets, reconstruction, predict, better, score, based, geometry, human, improve, test, multiple] [data, model, prior, true, bias, figure, approach, modeling, generative, likelihood, sampling, process, simulation, nonparametric] [skill, student, exercise, knowledge, bkt, technique, wcrp, nskill, factorization, synthetic, nexercise, matrix, occupied, correctly, nstudent, specific, tutoring, datashop, ntable, assumes, worse, intelligent, recovered, seated] [state, cognitive, single, restaurant, chinese, inferred, response, lfa, trial, pslc, dynamical, temporal] [strong, assumption, practice, term, domain, weighted, require]
Smoothed Gradients for Stochastic Variational Inference
Stephan Mandt, David Blei
Stephan Mandt, David Blei
[set, reduce, random, number, consider, queue] [algorithm, current, previous, allocation, draw] [learning, natural, word, test, entire] [variational, variance, bias, data, svi, full, parameter, inference, local, david, average, latent, nsf, global, wdn, minibatch, unbiased, zdn, averaging, model, lead, processing, york, estimate, expectation, predictive, explore, posterior, computation] [noisy, compute, smaller, vector, small, involves, introduce, reduction, faster] [topic, dirichlet, document, neural, lda, vocabulary, three, noise, multinomial] [stochastic, gradient, squared, smoothed, optimization, error, averaged, large, method, size, update, note, stored, biased, function, convergence, rate, machine, term, iteration, storage, tradeoff, effective, objective, sag, empirical]
A Filtering Approach to Stochastic Variational Inference
Neil Houlsby, David Blei
Neil Houlsby, David Blei
[optimum, set] [algorithm, adaptive, current, best, online, oracle, setting, sequence, static] [learning, computed, work, natural, original, better] [variational, svi, step, gaussian, kalman, posterior, variance, elbo, parallel, estimate, model, inference, parameter, bmf, bayesian, approximate, data, observation, bias, schedule, perspective, approach, distribution, latent, larger, filtering, tvf, freedom, nsf, process, log, react, infer, shift, figure, kosarak] [noisy, matrix, compute, robust, arxiv] [lda, tracking, noise, neural, time, single, dirichlet, directly, filter] [coordinate, stochastic, size, update, optimization, error, rate, constant, estimation, method, large, bayes, gradient, statistical, require]
Deep Recursive Neural Networks for Compositionality in Language
Ozan Irsoy, Claire Cardie
Ozan Irsoy, Claire Cardie
[binary, node, set, number, structure, leaf, tree, constructed] [previous, space, conference, good, best, observe] [deep, recursive, layer, hidden, sentiment, word, network, rnn, natural, depth, work, multiple, output, representation, language, learning, shallow, parse, input, table, supervised, rnns, phrase, architecture, task, great, meaning, stacking, training, performance, sentence, applied, activation, compositionality, net, christopher, outperform, yoshua, dropout, andrew, well, richard, semantic, similarity, cool, trained, charming, paragraph] [processing, figure, international, model, application, prediction, larger, allows] [vector, second, intermediate, small] [neural, recurrent, internal, single, change, response, weight, width, achieving] [nearest, structural, note, error]
Convolutional Neural Network Architectures for Matching Natural Language Sentences
Baotian Hu, Zhengdong Lu, Hang Li, Qingcai Chen
Baotian Hu, Zhengdong Lu, Hang Li, Qingcai Chen
[random, experiment, def, map, relation] [order, strategy, competitor, padding] [matching, sentence, convolutional, convolution, natural, language, feature, representation, architecture, learning, pooling, training, composition, proposed, network, trained, work, recursive, embedding, deep, input, task, output, original, word, better, propose, sliding, mlp, semantic, similarity, hidden, performance, paraphrase, based, rich, score, layer, test, illustrated, text, gating, adequately, mbed, report, ability] [model, figure, modeling, local, sequential, perform, naturally] [vector, second] [neural, length, interaction, hierarchical, individual, response, window] [machine, function, large, positive, example, essentially]
A Framework for Testing Identifiability of Bayesian Models of Perception
Luigi Acerbi, Wei Ji Ma, Sethu Vijayakumar
Luigi Acerbi, Wei Ji Ma, Sethu Vijayakumar
[interval, mapping, map, consider, discrete, number] [space, decision, probability, expected, case, optimal, design] [task, reference, reconstruction, visual, human, work] [model, prior, bayesian, measurement, estimate, likelihood, distribution, log, supplementary, posterior, true, mixture, process, approximate, allows, inference, figure, parameter, density, setup, priori] [analysis, stage, recovered, suggests, support] [observer, internal, experimental, noise, sensory, stimulus, perception, speed, response, motor, qprior, degeneracy, neurosci, reconstructed, sensorimotor, inferred, time, perceptual, nexp, short, magnitude, mental, pmeas, bsl, qmeas, range, nat, sensation, shared, behavior, change, timing, lapse, sci] [loss, function, assume, framework, estimation, large, class, theoretical, theory]
Algorithm selection by rational metareasoning as a model of human strategy selection
Falk Lieder, Dillon Plunkett, Jessica B. Hamrick, Stuart J. Russell, Nicholas Hay, Thomas Griffiths
Falk Lieder, Dillon Plunkett, Jessica B. Hamrick, Stuart J. Russell, Nicholas Hay, Thomas Griffiths
[metareasoning, rational, selection, sort, sorting, people, runtime, chose, cocktail, number, set, experiment, lagoudakis, random, comprised, inversely, pilot, execution, credible, mapping, list, solution, existing, apply, superior] [strategy, algorithm, choose, adaptive, expected, order, previous, berkeley, chosen, best, sequence, reinforcement, case] [performance, test, human, learning, learn, score, training, evaluated, long, predicted, input, evaluation, based, computer, voc, pattern, trained, performed, better] [model, figure, bayesian, posterior, regression, conditional] [randomly] [merge, cognitive, time, choice, psychological, three, sorted, length, estimated, mental, insertion, short] [method, problem, theory, linear, applying]
Decoupled Variational Gaussian Inference
Mohammad E. Khan
Mohammad E. Khan
[concave, set, number, linearly, existing, binary, maximum, apply, remains, maximizes] [lower, bound, algorithm, conference, max, result, case, good, clear] [computed, learning, work, based] [variational, inference, gaussian, bayesian, decoupled, data, approach, computation, likelihood, latent, model, supplementary, log, opper, posterior, vwn, distribution, approximate, covariance, global, parameterization, emtiyaz, mohammad, regression, chapter, reparameterization, international, process, prediction, converge, local] [compute, sparse, approximation, second, point, fast, compare, matrix] [time, range, nonlinear] [method, linear, lagrangian, function, convergence, large, term, gradient, journal, machine, optimization, convex, problem, form, cholesky, constrained, converges, note, respect, negative, dual, subproblem, iteration]
Analysis of Variational Bayesian Latent Dirichlet Allocation: Weaker Sparsity Than MAP
Shinichi Nakajima, Issei Sato, Masashi Sugiyama, Kazuho Watanabe, Hiroko Kobayashi
Shinichi Nakajima, Issei Sato, Masashi Sugiyama, Kazuho Watanabe, Hiroko Kobayashi
[map, solution, energy, number, corollary, collaborative, mechanism, variable] [case, lemma, allocation, appendix, proof, exp, previous] [learning, word, dense, based, hidden, compared] [bayesian, log, variational, latent, distribution, data, model, limit, figure, true, mixture, hyperparameters, posterior, markov, larger, prior, inference, stationary] [analysis, sparsity, matrix, sparse, leading, vector, approximation, denotes, factorization, support, requires, row] [topic, free, dirichlet, lda, behavior, estimated, document, maxh, pba, pbb, multinomial, minh, investigate, induce, berlin, tokyo, japan, watanabe, phenomenon, express, assumed, music, single, vocabulary] [asymptotic, theorem, term, assume, threshold, theoretical, size, occurrence, machine, theory, theoretically, consistency, journal, form, stochastic, large]
Probabilistic Differential Dynamic Programming
Yunpeng Pan, Evangelos Theodorou
Yunpeng Pan, Evangelos Theodorou
[belief, differential, programming, number, set] [policy, optimal, search, arm, algorithm, unknown, total] [learning, based, proposed, training, output, pair, joint, learned, computed, accurate, input] [local, gaussian, model, distribution, probabilistic, approach, predictive, uncertainty, step, data, kernel, bayesian, sampled, sample, target, variance, covariance, forward] [approximation, compute, requires, comparison, iterative] [pddp, control, state, time, ddp, pilco, trajectory, dynamic, system, physical, exk, controller, nominal, optimized, cdip, jacobian, classical, covf, nonlinear, varf, angular, velocity, inverted, indicate, xgoal, account] [cost, function, optimization, computational, locally, linear, framework, quadratic, problem, augmented, stochastic, machine, applying, method]
Algorithms for CVaR Optimization in MDPs
Yinlam Chow, Mohammad Ghavamzadeh
Yinlam Chow, Mohammad Ghavamzadeh
[consider, set, mathematical] [policy, algorithm, mdp, terminal, discounted, decision, optimal, appendix, probability, expected, spsa, mdps, current, critic, stopping, easy, incrementally, initial, agent, return, conference, case, alternative] [work, proposed, performance, level, original, propose] [estimate, parameter, var, markov, distribution, log, standard, unbiased, issue, conditional, generated, approach, variance] [approximation, incremental, point, vector, extended] [state, system, time, parameterized, trajectory] [gradient, function, loss, optimization, cvar, cost, risk, stochastic, update, augmented, objective, convergence, measure, note, method, lagrangian, min, estimator, linear, problem, tolerance, written]
Beta-Negative Binomial Process and Exchangeable Random Partitions for Mixed-Membership Modeling
Mingyuan Zhou
Mingyuan Zhou
[random, number, partition, set, marginal, paper] [probability, total, lemma] [word, representation, performance, joint] [process, model, distribution, gibbs, parameter, data, sampler, modeling, fully, sampling, latent, mixture, prior, nonparametric, bayesian, figure, prediction, inference, sample, likelihood, predictive] [column, vector, matrix, row] [bnbp, topic, exchangeable, dirichlet, count, eppf, njk, binomial, collapsed, beta, dependent, grouped, cluster, perplexity, expressed, heldout, rule, poisson, corpus, inferred, shared, observed, jth, governed, distinct, psyreview, kth, document, borel, jacm, multinomial, three, mtest, nvjk, hierarchical, chinese] [group, derive, function, size, negative, stochastic, measure, smoothing]
The Infinite Mixture of Infinite Gaussian Mixtures
Halid Z. Yerebakan, Bartek Rajwa, Murat Dundar
Halid Z. Yerebakan, Bartek Rajwa, Murat Dundar
[set, number, execution, structure, discrete, random] [base, version, total] [multiple, proposed, level, higher, layer, accurate, work, performed, predicted, table] [data, mixture, model, gmm, distribution, gaussian, sharing, colgibbs, gibbs, modeling, sampling, posterior, prior, inference, variational, sampler, process, covariance, predictive, sample, parameter, macro, allows, scaling, gaussians, generative, dependency, standard, average] [component, clustering, corresponding, matrix, formed] [cluster, dpmg, indicator, sck, dirichlet, single, skewed, individual, three, cell, lymphoma, time, micro, actual, ten, rare, modeled, flowcap, collapsed, thirty, produced, american, hyperspectral] [instance, bayes, group, journal, statistical]
Parallel Feature Selection Inspired by Group Testing
Yingbo Zhou, Utkarsh Porwal, Ce Zhang, Hung Q. Ngo, Long Nguyen, Christopher Ré, Venu Govindaraju
Yingbo Zhou, Utkarsh Porwal, Ce Zhang, Hung Q. Ngo, Long Nguyen, Christopher Ré, Venu Govindaraju
[scoring, number, selection, set, pfs, wrapper, subset, collection, binary, existing, fgm, identify, degree, high, paper, random, relation, madelon] [relevant, design, probability, total, result, best] [feature, test, testing, performance, based, accuracy, score, selected, select, learning, natural, extraction, datasets, original, report, higher, input, indicates, joint, adding, improves] [data, separable, approach, figure, scheme, parallel, estimate, step, conditional, supplementary, sequential, university, distribution] [running, small, matrix, exactly, condition, denotes, rank, vector, gisette, distance] [time, mutual, three, recall, stability] [method, function, large, group, machine, framework, department, divergence, bayes, theorem, positive, respect]
Spectral k-Support Norm Regularization
Andrew M. McDonald, Massimiliano Pontil, Dimitris Stamos
Andrew M. McDonald, Massimiliano Pontil, Dimitris Stamos
[set, ball, perturbation, cardinality, paper, number] [proposition, algorithm, case, optimal, result, setting, order, extend, study, chosen] [box, learning, invariant, table, unit, test, performance, report] [computation, parameter, log, data, figure] [norm, matrix, spectral, rank, proximity, operator, vector, trace, completion, jester, compute, special, movielens, low, small, singular, elastic, orthogonally, suggests, multitask, thresholding, absolute, support, recover, extended, numerical, equivalent] [cluster, noise, neural, real, von] [convex, regularization, problem, note, journal, error, dual, method, theorem, squared, machine, solve, statistical, gradient, optimization, proximal, inf, derive, linear, loss]
A Safe Screening Rule for Sparse Logistic Regression
Jie Wang, Jiayu Zhou, Jun Liu, Peter Wonka, Jieping Ye
Jie Wang, Jiayu Zhou, Jun Liu, Peter Wonka, Jieping Ye
[solution, set, number, ratio, identify, existing] [lemma, upper, bound, optimal, general, max, case, discarding, implies, algorithm, notice, admits] [feature, performance, proposed, based, table, novel, compared, learning, scale, newsgroup] [data, regression, parameter, university, demonstrate, figure, health] [sparse, running, vector, compare, matrix, comparison, small, product, special] [rule, time, state, real, science, recall] [slores, strong, screening, logistic, problem, regularized, safe, optimization, convex, inactive, solve, theorem, solving, dual, closed, estimation, rejection, solver, lasso, discarded, form, assume, large, applicable, arizona, discard, computational, method, duality, effective, lagrangian, multiplier, kkt, recreation, cost, yahoo, prostate, including]
Generalized Dantzig Selector: Application to the k-support norm
Soumyadeep Chatterjee, Sheng Chen, Arindam Banerjee
Soumyadeep Chatterjee, Sheng Chen, Arindam Banerjee
[set, consider, provide, ball, selection, exists, existing, root, number] [algorithm, general, search, bound, upper, order, focus, proposition, design] [unit, proposed, based, net, work, better, propose] [gaussian, log, standard, figure, parameter, conjugate, focused, larger, data, series, linearized] [norm, operator, recovery, analysis, vector, alternating, compute, cone, sparse, arindam, elastic, requires, suitable, matrix] [width, experimental, choice, length] [error, statistical, dantzig, proximal, prox, theorem, admm, estimation, selector, generalized, argmin, method, framework, inexact, note, optimization, dual, update, direction, journal, convex, computational, problem, roc, linear, solve, complexity, solving, accelerated, regularized, function, peter, proxic, example]
Exact Post Model Selection Inference for Marginal Screening
Jason D. Lee, Jonathan E. Taylor
Jason D. Lee, Jonathan E. Taylor
[selection, marginal, variable, set, proportion, interval, constructed, procedure, apply, pivotal, valid, paper, subset, construct, cover, cdf, applicability, diabetes, truncated, corollary, commonly, benedikt, hannes, lukas] [coverage, algorithm, adjusted, design, focus, jonathan, selects, methodology] [selected, work, matching, test, testing, dataset, jason, level] [model, distribution, conditional, regression, inference, figure, gaussian, standard, estimate, extremely, perform] [orthogonal, exact, matrix, omp, arxiv, preprint, condition, vector, pursuit, residual, element] [event, response, nominal, post, represent, described, noise, encode, snr] [screening, linear, framework, lasso, statistical, form, including, hypothesis, theorem, method, constrained, estimator, journal, assume, peter, annals, rate]
On Iterative Hard Thresholding Methods for High-dimensional M-Estimation
Prateek Jain, Ambuj Tewari, Purushottam Kar
Prateek Jain, Ambuj Tewari, Purushottam Kar
[hard, number, high, set, apply, existing] [algorithm, lemma, offer, general, result, appendix, satisfy, probability, previous, universal, setting, study] [level, work, learning] [restricted, regression, log, global, parameter, data, step, popular, variety] [sparsity, thresholding, matrix, sparse, recovery, analysis, greedy, iterative, support, condition, htp, projection, projected, rsc, foba, pursuit, singular, rip, noisy, iht, running, compressive] [property, noise, slower] [statistical, function, estimation, gradient, loss, linear, arg, convex, theorem, convergence, strong, dimensional, problem, large, method, error, descent, family, require, note, empirical, martin, differentiable, consistent]
Zero-shot recognition with unreliable attributes
Dinesh Jayaraman, Kristen Grauman
Dinesh Jayaraman, Kristen Grauman
[random, forest, tree, node, set, existing, left, label, fractional, recursively] [decision, gain, operating] [attribute, training, unseen, signature, learning, validation, object, discriminative, visual, train, dap, test, semantic, recognition, novel, idea, category, split, image, unreliability, trained, awa, unreliable, labeled, learn, predict, accuracy, shot, datasets, sun, scene, false, discovered, level, reliable, select, work, sig, solely] [data, approach, model, uncertainty, true, standard, full, figure, child, distribution, prediction, prior] [denote, vector] [noise, account, three, indicator] [class, method, roc, positive, error, large]
Articulated Pose Estimation by a Graphical Model with Image Dependent Pairwise Relations
Xianjie Chen, Alan L. Yuille
Xianjie Chen, Alan L. Yuille
[set, unary, number, graph, graphical, relation, random] [type, conference, space, detect, position] [image, pose, spatial, tij, human, flic, vision, computer, dataset, training, dcnn, pcp, lsp, pattern, recognition, buffy, tji, idpr, annotated, patch, score, learning, appearance, trained, predict, evaluation, table, articulated, wji, convolutional, learn, percentage, outperforms, pdj, pictorial, presence, datasets, test, body, labeled, based, conv, well, deeppose] [model, figure, local, standard, mixture, conditional, full, precision, international, data] [pairwise, strict, relational, detected] [dependent, weight, state, wij, neural, cluster] [method, relative, estimation, term, threshold, machine, positive, note]
Optimal rates for k-NN density and mode estimation
Sanjoy Dasgupta, Samory Kpotufe
Sanjoy Dasgupta, Samory Kpotufe
[mode, procedure, set, exists, consider, continuous, connected, surprisingly] [lemma, probability, optimal, unknown, bound, general, result, main, literature, setting] [work, level, neighborhood, salient, false, based] [density, sample, estimate, figure, equation, mass, kernel, allows, distribution, multivariate, nonparametric, approach, david] [estimating, denote, analysis, early] [event, cluster, single, direct] [estimation, smoothness, neighbor, assume, annals, assumption, practical, inf, pick, nearest, consistency, problem, suppose, theorem, continuity, arg, simple, estimator, gradient, journal, form, asymptotic, returned, statistical, machine, essentially, empirical, function, convergence, risk, consistent, rate]
Generalized Higher-Order Orthogonal Iteration for Tensor Decomposition and Completion
Yuanyuan Liu, Fanhua Shang, Wei Fan, James Cheng, Hong Cheng
Yuanyuan Liu, Fanhua Shang, Wei Fan, James Cheng, Hong Cheng
[core, set, existing, formulated, apply, solution] [algorithm, proof, appendix, optimal, lower, general] [proposed, propose, applied] [data, latent, model, scheme, missing, supplementary, scalable, sampling] [tensor, ghoi, schatten, matrix, decomposition, completion, minimization, wtucker, lrtc, rank, rse, recovery, falrtc, operator, denotes, smaller, analysis, unfolding, orthogonal, snm, tucker, wcp, running, robust, synthetic, alternating, singular, hooi, siam, factorization, hong, grant, point, square, faster, norm] [time, observed] [problem, method, convex, solving, theorem, min, optimization, computational, size, updating, generalized, convergence, solve, update, iteration, respect, kkt, admm, address, statistical, assume, function]
Neural Word Embedding as Implicit Matrix Factorization
Omer Levy, Yoav Goldberg
Omer Levy, Yoav Goldberg
[number, consider, solution, procedure] [context, optimal, algorithm, alternative, implicit, space, result, sense, probability, yoav, best] [word, association, better, training, similarity, semantic, well, nlp, pair, representation, embedding, work, performs, language, ecn, distributional, dense, trained, natural, based] [log, model, perform, sampled, equation, observation] [matrix, sgns, svd, pmi, sppmi, ppmi, shifted, factorization, factorizing, analogy, linguistic, vector, unobserved, syntactic, sparse, implicitly, frequent, exact, unigram, approximation, spectral, product, omer, levy] [observed, common, dimensionality, neural, mutual, presented, derived, perfect, corpus] [objective, method, negative, weighted, computational, stochastic, positive, metric, optimizing, gradient, measure, corresponds]
Active Learning and Best-Response Dynamics
Maria-Florina F. Balcan, Christopher Berlind, Avrim Blum, Emma Cohen, Kaushik Patnaik, Le Song
Maria-Florina F. Balcan, Christopher Berlind, Avrim Blum, Emma Cohen, Kaushik Patnaik, Le Song
[high, number, random, ball, label, consider, radius, move, consensus] [probability, algorithm, side, initial, result, bound, lemma, order, majority, setting, decision, exp, case, payoff, round, uniform, total] [learning, denoising, unit, work, learn, question, svm, correct, level, labeled] [target, figure, supplementary, local, true, sample] [distance, small, low, correctly, point, power, vector, greater, randomly, analysis, inside] [noise, fraction, state, experimentally] [active, sensor, update, error, separator, positive, asynchronous, negative, synchronous, theorem, large, agnostic, function, rate, linear, updated, generalization, passive, constant, theoretical, simple, incorrectly, incorrect, boundary, coordinate, applying, query, outperforming, georgia, method]
Provable Tensor Factorization with Missing Data
Prateek Jain, Sewoong Oh
Prateek Jain, Sewoong Oh
[random, number, solution, exists, tmax, provide, unique, consider, set] [algorithm, probability, proof, max, order, lemma, initial, optimal, result, bound, study, appendix, case] [higher] [log, step, sample, missing, estimate, propagation, sampled, scaling, local, approach] [tensor, matrix, completion, analysis, alternating, decomposition, minimization, orthogonal, rank, exact, norm, spectral, power, recovery, requires, recover, factorization, initialization, numerical, robust, small, singular, exactly, distance, tijk, denote, incoherence, approximation, required, provable, planted, symmetric] [connectivity, directly, observed] [theorem, method, error, problem, positive, constant, assume, prove, linear, size, iteration, generalization, dependence, function, convergence, iterates, machine, complexity]
Global Sensitivity Analysis for MAP Inference in Graphical Models
Jasper De Bock, Cassio P. de Campos, Alessandro Antonucci
Jasper De Bock, Cassio P. de Campos, Alessandro Antonucci
[map, robustness, instantiation, perturbation, set, pgms, facial, pgm, xik, probable, unique, consider, number, hamming, factor, discrete, perturb, mrf, candidate, variable, remains, equal, graphical, parametrized] [probability, best, algorithm, action, perturbed, case, max, general, considered] [accuracy, based, testing, work, predicted, joint, level, split, training, learning, applied, evaluating] [mass, data, inference, global, local, bayesian, model, computation, figure, conditional, probabilistic, true, approach, international] [second, vector, analysis, robust, compute, distance, compact, technique, corresponding] [single, sensitivity, time, qualitative, state] [respect, function, measure, problem, consistent, complexity, instance, family, theorem, example, queried, optimization]
Multi-scale Graphical Models for Spatio-Temporal Processes
firdaus janoos, Huseyin Denli, Niranjan Subrahmanya
firdaus janoos, Huseyin Denli, Niranjan Subrahmanya
[graphical, structure, set, proportional, variable, solution, high, number, mixed] [appendix, order, upper, exploration, diffusion] [learning, spatial, multiple, ground, network, spatially, ieee] [model, global, var, local, data, latent, prior, supplemental, process, concentration, parameter, approach, step, larger, inversion] [vector, component, estimating, sparse, small, operator, approximation, distance] [system, time, physically, observed, pressure, derived, tracer, connectivity, clet, estimated, impulse, dynamical, directly, velocity, permeability, indicate, injected, exhibit] [group, problem, lasso, estimation, large, theorem, constant, method, regularization, proximal, solving]
Distance-Based Network Recovery under Feature Correlation
David Adametz, Volker Roth
David Adametz, Volker Roth
[marginal, number, cancer, graphical, edge, maximum, degree, arbitrary, dna] [algorithm, correlated, goal, probability] [feature, network, false, based, better] [likelihood, latent, tiwnet, timt, log, bayesian, model, figure, prior, posterior, full, data, inference, approach, covariance, gene, mismatch, gaussian, trcm, acceptance, arrive, parameter, fully, variance, true, dij, highly, process, pathway, bhattacharyya, mcmc, inherent, hyperparameters, scheme, wishart] [matrix, distance, synthetic, pairwise, second, recovery, estimating] [independent, classical, correlation, biological, inverse, choice, observed, depicted, gamma, american] [statistical, journal, positive, regularization, loss, assumption, require, note]
Decomposing Parameter Estimation Problems
Khaled S. Refaat, Arthur Choi, Adnan Darwiche
Khaled S. Refaat, Arthur Choi, Adnan Darwiche
[variable, number, decomposed, set, maximum, structure, consider, compatible, induced, partition, graphical, leaf, complete, corollary, appear, uai, subset, random] [algorithm, conference, called, evidence, appears, case] [network, dataset, learning, hidden, proposed, datasets, based, employ] [likelihood, bayesian, stationary, parameter, inference, figure, decomposing, data, soundness, water, adnan, alarm, incomplete, missing, engine, andes, diagnose, chain, edml, uncertainty, arthur, smile, khaled] [component, decomposition, technique, point, denote, projected, smaller] [observed, time, three, independent, distinct, count] [function, problem, example, convergence, boundary, size, threshold, computational, iff, gradient]
Scaling-up Importance Sampling for Markov Logic Networks
Deepak Venugopal, Vibhav G. Gogate
Deepak Venugopal, Vibhav G. Gogate
[number, set, variable, asymptotically, marginal, symbolic, ratio, subset, scalability] [evidence, conference, algorithm, counting, probability] [ground, based, network, accuracy, computed, work] [sample, inference, sampling, markov, proposal, distribution, true, approach, unbiased, sampler, probabilistic, estimate, gibbs, scalable, step, generate, figure, university, illustrate, approximate] [compute, computing, approximation, corresponding, clustering, relational, compressed, knowledge] [weight, cluster, time] [mln, formula, lifted, grounding, complexity, logical, mlns, logic, large, assume, query, weighted, predicate, example, atom, symbol, note, domain, error, outer, normal, corresponds]
Sparse Random Feature Algorithm as Coordinate Descent in Hilbert Space
En-Hsu Yen, Ting-Wei Lin, Shou-De Lin, Pradeep K. Ravikumar, Inderjit S. Dhillon
En-Hsu Yen, Ting-Wei Lin, Shou-De Lin, Pradeep K. Ravikumar, Inderjit S. Dhillon
[random, solution, number, set, corollary, linearly, map, program] [algorithm, optimal, space, minimizing, bound, guarantee, result, minimize, decision, uniform] [feature, training, proposed, learning, achieved, testing] [kernel, approach, distribution, prediction, model, data, comparable, regression, standard, approximate, step, university, gaussian, sampling, larger] [basis, sparse, approximation, support, greedy, laplacian, hilbert, vector, fourier, analysis, decomposition, small, compare, corresponding, exact, compact] [time, memory, derivative] [boosting, coordinate, descent, loss, function, convergence, method, problem, randomized, min, perceptron, objective, error, theorem, ntr, ttr, size, note, rbf, working, iteration, rate, machine, term, converges, solve, wref, argmin, solver, large, linear]
A framework for studying synaptic plasticity with neural spike train data
Scott Linderman, Christopher H. Stock, Ryan P. Adams
Scott Linderman, Christopher H. Stock, Ryan P. Adams
[potential, number, map, connected, set, random, wmax] [static, order, probability, algorithm, design, study] [learning, network, train, work, optical, proposed] [model, particle, inference, likelihood, figure, log, data, standard, distribution, conditional, chain, mcmc, bayesian, sampling, processing, sample, process, true, predictive, approach, markov, prior] [additive] [synaptic, weight, stdp, spike, neural, plasticity, rule, glm, time, neuron, multiplicative, nonlinear, excitatory, impulse, synapse, dynamical, nature, biophysical, spiking, trajectory, three, dynamic, biological, undergoing, inferred, postsynaptic, membrane, distinguish, experimental, noise, stimulus] [linear, function, rate, statistical, framework, generalized, journal, computational]
Global Belief Recursive Neural Networks
Romain Paulus, Richard Socher, Christopher D. Manning
Romain Paulus, Richard Socher, Christopher D. Manning
[node, tree, belief, root, number, set, structure, classify] [context, contextual, order, best] [word, sentiment, recursive, phrase, training, rnn, labeled, task, semantic, unsupervised, dataset, learning, twitter, trained, semeval, better, feedbackward, sentence, bidirectional, dropout, well, additional, hybrid, correct, natural, table, compositional, deep, train, wlf, wrf, idea, performs, language, performance, purely, bad, android, supervised, composed, work, feature, representation, wrb, combination, softmax, network, wlb] [model, global, step, standard, backward, figure, forward, international, computation, data] [vector, comparison, analysis, introduce, matrix] [neural, feedforward, recurrent, competition, single, time] [example, positive, large]
Deep Networks with Internal Selective Attention through Feedback Connections
Marijn F. Stollenga, Jonathan Masci, Faustino Gomez, Jürgen Schmidhuber
Marijn F. Stollenga, Jonathan Masci, Faustino Gomez, Jürgen Schmidhuber
[map, number, set] [policy, feedback, action, focus, reinforcement, search, agent] [output, image, maxout, layer, network, convolutional, dasnet, deep, input, attention, learning, trained, pooling, visual, performance, recognition, object, selective, natural, softmax, snes, activation, feature, cnns, work, learn, training, computer, improve, cnn, net, cat, era, correct, learned, vision, parameterised] [processing, observation, step, pass, figure, model, standard, allowing, distribution, computation, perspective, sequential] [vector, small, second] [neural, feedforward, evolution, internal, change, time, dimensionality, control, recurrent, state, attentional, direct] [size, class, journal]
Iterative Neural Autoregressive Distribution Estimator NADE-k
Tapani Raiko, Yao Li, Kyunghyun Cho, Yoshua Bengio
Tapani Raiko, Yao Li, Kyunghyun Cho, Yoshua Bengio
[number, set, structure, consider] [conference, probability, case, criterion, version, best, previous] [training, deep, hidden, learning, performance, boltzmann, proposed, test, trained, better, well, table, reconstruction, train, computed, input, learned, denoising, network, outperforms, mnist, validation, based, original] [model, inference, nade, xobs, missing, xmis, log, distribution, variational, data, autoregressive, probabilistic, conditional, figure, generative, international, restricted, approach, markov, generate, mixture, density, step, mask] [compute] [neural, rbm, single, observed, weight, variability, independent] [machine, stochastic, gradient, estimator, iteration, regularization, generalized, empirical]
Extreme bandits
Alexandra Carpentier, Michal Valko
Alexandra Carpentier, Michal Valko
[maximum, consider, number, set, limited, provide, john] [extreme, pareto, arm, max, regret, bandit, bound, algorithm, xtreme, setting, order, anomaly, unter, optimal, probability, lemma, expected, oracle, tail, upper, negligible, heaviest, result, best, main, chet, hresholda, learner, scent, limiting, strategy, lower, allocate, highest, implies, focus] [network, detection, compared, learning, work, performance, ieee, performs] [distribution, step, data, expectation, larger, parametric, source, xit, sampling, figure, heavy] [second, exact, largest] [time, classical, event] [theorem, assumption, objective, depends, function, active, problem, prove, machine, converges]
Stochastic Multi-Armed-Bandit Problem with Non-stationary Rewards
Yonatan Gur, Assaf Zeevi, Omar Besbes
Yonatan Gur, Assaf Zeevi, Omar Besbes
[number, adversarial, set, admissible, structure] [regret, variation, reward, policy, expected, best, mab, order, arm, budget, action, achievable, proof, horizon, oracle, sequence, setting, bandit, decision, studied, optimal, allocation, minimax, establishing, optimality, changing, lower, bound, environment, main, restless, static, established, benchmark, allowed, guarantee, unknown, incur, online] [performance, learning, achieve, compared, work] [log, stationary, university, distribution, step, sequential, allows, average] [denote, analysis, introduced, absolute, second] [dynamic, single, change, time, temporal, independent] [stochastic, theorem, class, batch, relative, rate, problem, complexity, epoch, function, linear, journal, theory, size, min, note, respect, constant, objective]
Using Convolutional Neural Networks to Recognize Rhythm Stimuli from Electroencephalography Recordings
Sebastian Stober, Daniel J. Cameron, Jessica A. Grahn
Sebastian Stober, Daniel J. Cameron, Jessica A. Grahn
[number, increase, classify, structure] [best, study, good] [learning, accuracy, convolutional, input, deep, table, layer, training, cnns, applied, performance, based, impact, test, validation, network, channel, human, cnn, learned, work, combination] [data, processing, bayesian, parameter, kernel, figure] [confusion, fast, spectrum, waveform] [eeg, rhythm, classification, frequency, east, african, neural, subject, three, rhythmic, western, individual, length, hop, bar, single, brain, auditory, beat, offset, trial, perception, music, width, musical, time, perceived, stimulus, xxxx, xxx, presented] [size, structural, rate, optimization, machine, journal]
Neurons as Monte Carlo Samplers: Bayesian Inference and Learning in Spiking Networks
Yanping Huang, Rajesh P. Rao
Yanping Huang, Rajesh P. Rao
[number, sum, random, release, proportional, discrete] [probability, transition, initial, online, previous, expected, algorithm] [learning, hidden, network, based, proposed, layer, represented, performance] [model, inference, distribution, posterior, bayesian, figure, monte, carlo, true, observation, markov, prediction, estimate, variance, sampling, equation, standard, likelihood, probabilistic, volatility, supplementary, approximate] [approximation, matrix, noisy, involves] [neural, time, recurrent, spiking, sensory, state, neuron, spike, emission, population, wij, represents, synaptic, weight, variability, cortical, activated, poisson, observed, mij, feedforward, neuronal, represent, preferred, activity, solid, noise, rule] [convergence, stochastic, estimator, rate, form, updated]
Variational Gaussian Process State-Space Models
Roger Frigola, Yutian Chen, Carl Rasmussen
Roger Frigola, Yutian Chen, Carl Rasmussen
[procedure] [transition, lower, optimal, bound, evidence, future, order, online, algorithm, depend, goal] [learning, test, learn, training, based, natural, ability, conventional, work] [variational, distribution, gaussian, model, posterior, approach, latent, inducing, process, inference, bayesian, approximate, prior, data, parametric, log, sequential, monte, series, sample, tractable, covariance, particle, auxiliary, carlo, processing, sampling, limit, prediction, international, figure, likelihood, pmcmc, predictive, elbo] [sparse, second] [nonlinear, time, dynamical, state, system, neural, spike, length, independent, panel, control] [smoothing, function, stochastic, linear, respect, cost, machine, min, method, problem, note, complexity]
Automated Variational Inference for Gaussian Process Models
Trung V. Nguyen, Edwin V. Bonilla
Trung V. Nguyen, Edwin V. Bonilla
[binary, set] [lower, bound, expected, general, deterministic] [learning, performance, computed, dataset, box, training] [variational, gaussian, inference, log, likelihood, latent, full, covariance, mixture, regression, process, posterior, mix, sampling, distribution, agp, standard, predictive, warped, automated, nlpd, hyperparameters, elbo, figure, gpr, analytical, model, wgp, wrt, gaussians, approximate, gps, analytically, sse, approach, comparable, cox, bayesian, supplementary, monte, data, mcmc, hmc, creep, david, edward, carlo, equation] [exact, diagonal, approximation, faster, component, compare, factorization, univariate] [time, magnitude] [method, function, term, family, practical, note, theorem, error, framework, usps, machine]
Bandit Convex Optimization: Towards Tight Bounds
Elad Hazan, Kfir Levy
Elad Hazan, Kfir Levy
[set, adversary, paper, limited, consider, interior, remains, ball] [regret, algorithm, lemma, online, bco, bandit, barrier, setting, case, player, version, bound, decision, optimal, main, dikin, receiving, expected, sequence, exploration, regreta, upper, ellipsoid, chooses, choose, oco, best, ensures, general, previous, choosing, bounded, proof, feedback, elad, result, lower] [understanding, unit, work, learning, well] [sampling, log, equation, scheme, unbiased, estimate, full] [point, noisy, norm, matrix, denote, special] [time] [convex, loss, function, gradient, smoothed, optimization, smooth, linear, method, framework, arg, theorem, respect, rate, constrained, note, assume, stochastic, cost, positive, machine, solving, achieves]
Gaussian Process Volatility Model
Yue Wu, José Miguel Hernández-Lobato, Zoubin Ghahramani
Yue Wu, José Miguel Hernández-Lobato, Zoubin Ghahramani
[asymmetric, number, high] [transition, previous, algorithm, expected, main, return, space, initial] [hidden, learning, learn, training, table, learned, performance, predicted, dataset, test, compared] [rapcf, model, gaussian, volatility, particle, process, garch, data, pgas, egarch, predictive, inference, series, variance, figure, standard, average, posterior, parameter, log, relationship, covariance, observation, prediction, prior, equity, plot, likelihood, sampling, surface, audusd, forward, auxiliary, bayesian, rapf, ancestor, computationally, daily] [corresponding, rank, review, comparison] [time, state, nonlinear, real] [function, functional, method, positive, negative, large, journal, batch, statistical, linear, empirical]
Nonparametric Bayesian inference on multivariate exponential families
William R. Vega-Brown, Marek Doniec, Nicholas G. Roy
William R. Vega-Brown, Marek Doniec, Nicholas G. Roy
[maximum, set, entropy, structure, formulated, provide] [choose, unknown, exchange, case] [based, input, training, performance, predicted, table] [distribution, likelihood, gaussian, model, process, inference, kernel, prior, covariance, wishart, bayesian, exponential, heteroscedastic, posterior, latent, approach, multivariate, predictive, target, regression, perform, approximate, local, equation, figure, data, conjugate, gwp, true, generate, kniw, generative, observation, evaluate, infer, uninformative, mll, periodic, parameter, allow, nonparametric, motorcycle] [exact, extended, compare, running, comparison] [inferred, time, inverse, described, independent] [function, note, divergence, family, assumption, stochastic, generalized, method, weighted, estimation, smoothness, journal, statistical, satisfying, problem, implemented, positive]
Improved Multimodal Deep Learning with Variation of Information
Kihyuk Sohn, Wenling Shang, Honglak Lee
Kihyuk Sohn, Wenling Shang, Honglak Lee
[map, set, number, random, provide, binary] [good, variation, transition, algorithm] [learning, multimodal, deep, training, joint, minvi, network, modality, multiple, visual, proposed, input, performance, text, recognition, trained, boltzmann, feature, database, mrbm, train, wjk, image, learn, representation, test, pascal, mdrnn, layer, improvement, table, wik, retrieval, propose, unsupervised, based, report, voc, lvi, contrastive] [data, model, distribution, conditional, generative, missing, equation, target, restricted, log, inference, figure] [] [neural, encoding, recurrent, rbm, shared, visible, derived] [objective, unimodal, query, stochastic, gradient, note, theorem, function]
Mind the Nuisance: Gaussian Process Classification using Privileged Noise
Daniel Hernández-lobato, Viktoriia Sharmanska, Kristian Kersting, Christoph H. Lampert, Novi Quadrianto
Daniel Hernández-lobato, Viktoriia Sharmanska, Kristian Kersting, Christoph H. Lampert, Novi Quadrianto
[binary, set, label, discovery, factor, consider] [best, conference, easy, associated, evidence] [privileged, learning, training, dataset, svm, test, original, performance, slope, additional, based, proposed, deep, visual, attribute, performs, awa, decaf, input, semantic, vapnik, work, performed] [data, gaussian, latent, process, likelihood, gpc, prior, model, average, prediction, heteroscedastic, standard, approximate, posterior, kernel, distribution, probit, figure, predictive, inference, expectation, propagation, bayesian, nuisance, uncertainty, regression, processing, international, xnew, austria, sample, parameter, qold] [rank, corresponding, noisy] [noise, neural, system] [function, machine, method, term, error, statistical, form, respect, domain, assume, strong]
Fast Kernel Learning for Multidimensional Pattern Extrapolation
Andrew Wilson, Elad Gilboa, John P. Cunningham, Arye Nehorai
Andrew Wilson, Elad Gilboa, John P. Cunningham, Arye Nehorai
[structure, number, marginal, runtime, discovery, region] [grid, total, considered, alternative] [pattern, training, learning, test, scale, image, learned, automatically, input, learn, datasets, fig, train, scene, long, accuracy] [kernel, gpatt, inference, gaussian, data, extrapolation, ssgp, figure, missing, process, multidimensional, mixture, expressive, smp, gps, stress, inducing, scalable, model, fitc, standard, kronecker, likelihood, inpainting, plate, tread, perform, log, nonparametric, extrapolate, metal, sophisticated, scaling, regression, popular, processing, smse, university, hyperparameters, kper, full, approach, imaginary, approximate] [basis, fast, spectral, exact, product, small, compare, matrix, sparse] [range, neural, speed] [large, function, machine, method, complexity]
Multi-Scale Spectral Decomposition of Massive Graphs
Si Si, Donghyuk Shin, Inderjit S. Dhillon, Beresford N. Parlett
Si Si, Donghyuk Shin, Inderjit S. Dhillon, Beresford N. Parlett
[graph, number, label, structure, random, recommendation, hierarchy, quality, partition, recommender, apply] [algorithm, appendix, social, termination, good, version] [level, performance, computed, learning, table, achieve, dataset, based, better, proposed, datasets, compared, network] [approximate, figure, propagation, inductive] [mseigs, matrix, clustering, lanczos, spectral, eigenvectors, principal, decomposition, compute, eigs, rsvd, faster, eigenpairs, dominant, sparse, cosine, subspace, blklan, approximation, aii, eigendecomposition, smaller, divide, diagonal, propack, livejournal, small, conquer, rank, siam, aloi, early, condmat, imc, symmetric, corresponding, svd, consists] [time, single] [block, method, convergence, machine, randomized, theorem, form, problem, large, suppose]
Spectral Learning of Mixture of Hidden Markov Models
Cem Subakan, Johannes Traa, Paris Smaragdis
Cem Subakan, Johannes Traa, Paris Smaragdis
[number, permutation, mapping, structure, experiment, ratio] [algorithm, transition, sequence, lemma, initial, total, good] [learning, hidden, accuracy, based, proposed, presence, reconstruction, unit, work] [mhmm, global, hmm, markov, latent, model, data, mixture, figure, approach, observation, distribution, hmms, parameter, likelihood, stationary, gaussian, computationally, variance, true, mixing, amount, equation, third, estimate, moment] [spectral, matrix, eigenvalue, clustering, diagonal, second, mom, notation, corresponding, depermutation, longevity, largest, noisy, denote, condition, estimating, top] [state, cluster, noise, time, estimated, emission, magnitude, individual, initialized, length, determines] [method, estimation, block, machine, note, large, rate]
Large-Margin Convex Polytope Machine
Alex Kantchelian, Michael C. Tschantz, Ling Huang, Peter L. Bartlett, Anthony D. Joseph, J. D. Tygar
Alex Kantchelian, Michael C. Tschantz, Ling Huang, Peter L. Bartlett, Anthony D. Joseph, J. D. Tygar
[polytope, number, entropy, assignment, assigned, binary, set, maximum, high, random, sum, collection, smallest, argmaxk] [algorithm, conference, maximizing, total, max, good, space, result, bound, focus, worst, case] [training, learning, testing, dataset, based, propose, performance, accuracy, mnist, scale, train, work, perfectly, achieve] [data, local, figure, model, kernel, approach, larger, parameter, separable] [running, small, point, vector] [single, time, subject, nonlinear] [positive, margin, convex, linear, problem, cpm, error, large, machine, optimization, function, gradient, min, stochastic, sgd, negative, empirical, descent, solving, term, regularization, generalization, amm, unadj, computational, rate, conjecture, adjust, separator, corresponds, dimensional]
Feature Cross-Substitution in Adversarial Classification
Bo Li, Yevgeniy Vorobeychik
Bo Li, Yevgeniy Vorobeychik
[adversarial, adversary, set, evasion, number, spam, sma, solution, malicious, enron, constraint, lowd, naive, meek, evade, zaj, selection, acm, consider, sigkdd, attack, uci, binary, milp, making, threat, ebc] [game, algorithm, budget, learner, stackelberg, heuristic, security, conference, strategy, offer, general, open] [feature, learning, proposed, training, traditional, performance, based, test, select, word, natural, evaluation] [data, figure, model, approach, international, observation, highly] [reduction, corresponding, knowledge, comparison, vector, running, second, matrix] [response, phenomenon] [cost, machine, problem, linear, function, equivalence, optimization, query, class, instance, error, tradeoff, min, ideal, suppose, method]
Multi-Resolution Cascades for Multiclass Object Detection
Mohammad Saberian, Nuno Vasconcelos
Mohammad Saberian, Nuno Vasconcelos
[binary, set, high, number, structure] [algorithm, policy, adaptive, design, previous, learner] [cascade, detection, multiclass, mres, detector, learning, car, object, false, accuracy, mcboost, training, multiple, yzi, accurate, sbboost, learn, ieee, performance, computer, pattern, codeword, vision, driven, image, learned] [bias, figure, target, parallel, data, standard, computationally, log] [early, stage, low, small, intermediate, fast, analysis] [evolution, desired] [class, boosting, rate, complexity, margin, structural, weak, positive, sign, weighting, negative, example, large, arg, min, biased, predictor, computational, machine, late, denoted, respect, problem, implemented]
A Boosting Framework on Grounds of Online Learning
Tofigh Naghibi Mohamadpoor, Beat Pfister
Tofigh Naghibi Mohamadpoor, Beat Pfister
[set, entropy, number, fact, consider] [algorithm, proof, lemma, online, version, probability, bound, case, round, result, interesting, bounded] [learning, work, replacing, training, output, train, level, accuracy, based, viewpoint, proposed] [distribution, sample, approximate, approach] [projection, norm, vector, second, provable, simplex, sparsity, generalize, sparse, exact, projected, operator] [weight, choice, desired, derived, directly, noise, generates] [boosting, update, convex, hypothesis, bregman, theorem, lazy, weak, function, smooth, loss, madaboost, framework, active, inequality, error, agnostic, mirror, negative, maboost, descent, variant, sparseboost, term, divergence, sign, respect, dual, journal, gradient, min, note, generalized, arg, problem, family, duality, machine, learnability, margin, algorithmic, applying, assume]
Robust Logistic Regression and Classification
Jiashi Feng, Huan Xu, Shie Mannor, Shuicheng Yan
Jiashi Feng, Huan Xu, Shie Mannor, Shuicheng Yan
[random, programming, binary, consider, ratio, number, label, preprocessing, maximizes, adversarial, robustness, variable] [bound, probability, lemma, algorithm, guarantee, bounded, max, space, minimizing, result, maximizing, difference] [performance, training, proposed, based, output, truth, ground] [regression, parameter, sample, model, log, standard, generated, covariate, larger, figure, gaussian, distribution, expectation, data, concentration, supplementary, prediction] [robust, corrupted, outlier, noiseless, estimating, matrix, dimension, sparse, denote] [correlation, population, described, fraction, magnitude, noise] [rolr, logistic, linear, function, error, problem, loss, risk, empirical, authentic, large, estimation, suppose, inlier, inliers, positive, complexity, measure, solve, theorem, journal, statistical, solving, trimmed, achieves, constant, convex, department, assume, sign]
Multi-Class Deep Boosting
Vitaly Kuznetsov, Mehryar Mohri, Umar Syed
Vitaly Kuznetsov, Mehryar Mohri, Umar Syed
[set, tree, binary, mapping, maximum, sum] [base, bound, appendix, algorithm, max, decision, best, upper, admits, case, associated, guarantee] [learning, table, ensemble, based, deep, test, complex, performance, selected, reported, depth] [log, exponential, parameter, mixture, standard] [denote, vector] [presented, choice, derived] [convex, error, boosting, fsum, fmax, function, rademacher, logistic, hypothesis, theorem, complexity, loss, optimization, margin, family, generalization, objective, fcompsum, fmaxsum, mdeepboost, term, size, deepboost, mdeepboostsum, adaboost, convergence, problem, large, example, machine, consistency, empirical, coordinate, differentiable, risk, prove, koltchinskii, min, admit, direction, descent, regularized, annals, assume, structural, weighted, mdeepboostcompsum]
Spectral Methods for Supervised Topic Models
Yining Wang, Jun Zhu
Yining Wang, Jun Zhu
[number, variable, collection, movie] [algorithm, order, appendix, observable, bound, previous, search] [supervised, hybrid, based, learning, word, work, performance, proposed, training, accuracy, novel, reconstruction] [model, regression, latent, gibbs, sampling, distribution, sample, parameter, moment, prediction, mixing, local, prior, data, likelihood, modeling, perform, sampled, step, true, demonstrate, inference] [tensor, spectral, decomposition, recover, matrix, slda, power, recovery, orthogonal, vector, robust, denote, synthetic, review, analysis, simultaneous, diagonalization, norm] [topic, time, lda, underlying, dirichlet, reconstructed, response, estimated, document] [linear, method, complexity, error, empirical, large, term, update, problem, machine, practical, assume]
Spectral Methods for Indian Buffet Process Inference
Hsiao-Yu Tung, Alex J. Smola
Hsiao-Yu Tung, Alex J. Smola
[number, set, binary, random, structure, factor, expression, polynomial] [order, algorithm, goal, bounded, space, probability, alternative, lemma, conference] [learning, reconstruction, hidden, feature, accuracy] [latent, model, ibp, data, variational, inference, indian, concentration, buffet, approach, figure, process, infer, third, gene, generated, mcmc, university, kernel, tensorial, modeling, sampling] [spectral, tensor, analysis, power, diag, matrix, denote, sparse, row, additive, fourth, second, symmetric, nonzero, provided, compute, suitable, recover] [independent, noise, inferring, signal, zij, derived, dirichlet, correlation, directly, tool, inferred] [method, statistical, measure, note, machine, linear, journal, excess, derive, size]
A Representation Theory for Ranking Functions
Harsh H. Pareek, Pradeep K. Ravikumar
Harsh H. Pareek, Pradeep K. Ravikumar
[set, partially, consider, random, discrete, decomposed, permutation, sum, continuous] [ranking, listwise, case, base, general, result, reranking, reranked, order, analytic, conference, appendix, proposition, finetti, space, probability, ohsumed, decision, called, bounded, affect] [learning, representation, evaluation, feature, object, score, output, letor, diverse, retrieval, similarity, lack, applied, represented] [multivariate, distribution, international] [symmetric, tensor, decomposition, rank, pointwise, corresponding, analysis, compact, product, ndcg, spectral, vector, operator, multilinear, pairwise, refer, matrix, component] [exchangeable, document, reader, three] [theorem, function, theory, loss, functional, class, key, gradient, linear, corresponds, convex, query, method, note, functionals, term, require, computational]
Partition-wise Linear Models
Hidekazu Oiwa, Ryohei Fujimaki
Hidekazu Oiwa, Ryohei Fujimaki
[partition, number, region, set, map, energy, selection, easily] [bound, algorithm, space, best] [learning, feature, structured, table, better, input, datasets, performed, well, dataset, indicates, internet, performance, applied, test, work, propose] [model, local, global, regression, kernel, data, figure, approach, generated, prediction, standard, full, developed] [vector, residual, decomposition, second, formulation, analysis, fast] [weight, individual, represent] [linear, convex, proximal, optimization, locally, loss, activeness, generalization, respect, function, denoted, error, problem, active, effective, ldkl, empirical, method, complexity, derive, adp, computational, journal, large, xor, gradient, min, rademacher, logistic, theorem, census, summarizes, cost, machine]
PAC-Bayesian AUC classification and scoring
James Ridgway, Pierre Alquier, Nicolas Chopin, Feng Liang
James Ridgway, Pierre Alquier, Nicolas Chopin, Feng Liang
[set, qij, marginal, selection, random, structure, consider, dna, bipartite, scoring] [probability, exp, algorithm, van, ranking, general, bound, depend, version, choose] [score, feature, dataset, better, based, table, natural] [gaussian, prior, qqq, approach, bayesian, smc, log, monte, supplement, mcmc, target, distribution, carlo, slab, density, data, standard, figure, qqqqq, alquier, approximate, advantage, expectation, pima, rankboost, site, rij, sequential, posterior, normalising, covariates, crest, robbiano, sample, fij] [approximation, product, leading, sparse, second, extra] [auc, spike, independent, choice, implement] [class, qqqq, respect, empirical, constant, inf, assume, positive, derive, linear, note, roc, negative, rate, theorem, risk, regularization]
Extracting Certainty from Uncertainty: Transductive Pairwise Classification from Pairwise Similarities
Tianbao Yang, Rong Jin
Tianbao Yang, Rong Jin
[label, number, set, random, subset, complete, construct] [algorithm, space, proof, order] [similarity, proposed, labeled, learning, performance, based, feature, well, reconstruction, accuracy, indicates, utilize] [data, kernel, estimate, step, sampled, full, figure, true, application] [matrix, pairwise, clustering, spectral, coherence, top, column, provided, transductive, completion, skl, tpcmc, small, recover, uniformly, analysis, spanned, low, orthogonal, second, rank, recovered, belong, corresponding, denote, splice, recovery, completed, largest, gisette, introduce, randomly] [observed, three, estimated, cluster, perfect] [problem, measure, theorem, theoretical, simple, class, error, normalized, theory, optimization, large, method]
Learning Shuffle Ideals Under Restricted Distributions
Dongqu Chen
Dongqu Chen
[concept, polynomial, random, consider, collection, set, exists] [algorithm, space, probability, general, lower, lemma, bound, oracle, difference, appendix, result, main, case] [learning, language, labeled, answer, natural, pattern, accuracy] [distribution, kernel, generated, conditional, parameter, estimate, model, advantage, restricted] [principal, product, denote, greedy, exact, extended, condition, element, recovered] [independent, length, time] [string, query, statistical, class, pac, instance, ideal, tolerance, symbol, example, regular, augmented, problem, learnable, learnability, generalization, computational, markovian, theoretical, identical, positive, machine, constrained, simple, theorem, unrestricted, error, unimodal, feasibility, strong, journal, method, prove, negative, complexity, unimodality, theory, size, legitimacy, empirical, concatenation]
Augmentative Message Passing for Traveling Salesman Problem and Graph Partitioning
Siamak Ravanbakhsh, Reihaneh Rabbany, Russell Greiner
Siamak Ravanbakhsh, Reihaneh Rabbany, Russell Greiner
[message, passing, number, tsp, modularity, set, graph, random, salesman, subtour, constraint, traveling, community, augmentation, solution, factor, node, cutting, clique, edge, degree, marginal, belief, connected, procedure, enforce, adjacent, enull, partitioning, trend, ratio, programming, feasible, quality, smallest, tour, binary] [minimum, benchmark, optimal, optimality, combinatorial, current, start, order, context, general] [idea, slope, applied] [null, model, local, figure, full, sampling, incoming, inference, fully, approach] [distance, sparse, clustering, approximation, second, review, denote, exact, small, column] [time, three, physical, weight] [optimization, large, method, problem, linear, cost, plane, normalized, form, iff, weighted, negative, violated]
Projective dictionary pair learning for pattern classification
Shuhang Gu, Lei Zhang, Wangmeng Zuo, Xiangchu Feng
Shuhang Gu, Lei Zhang, Wangmeng Zuo, Xiangchu Feng
[number, set, label, code, energy] [algorithm, action, best] [dpl, training, representation, coding, learning, discriminative, proposed, recognition, learn, testing, face, yaleb, database, pair, image, discrimination, projective, accuracy, based, trained, pattern, reconstruction, table, structured, ieee, fddl, object, learned, conventional, visual, work, better, test, svm, jointly, feature, propose, performance, crc, learns, costly] [model, sample, data, parameter, advantage, analytical] [dictionary, sparse, analysis, extended, matrix, residual, sparsity, small, dimension, power, vector, faster, diagonal, reconstruct] [synthesis, time, signal, competing, experimental, represent] [class, arg, function, linear, update, method, min, optimization, convergence, query, complexity, achieves, problem, updating, objective]
Biclustering Usinig Message Passing
Luke O'Connor, Soheil Feizi
Luke O'Connor, Soheil Feizi
[biclustering, bcmp, biclusters, isa, number, overlapping, bipartite, penalty, cij, message, lij, expression, binary, graph, high, misclassified, edge, maximum, existent, iteratively, passing, ratio] [algorithm, tuples, search, case, optimal, proposition, total, heuristic] [computed, level, overlap, applied, performed] [model, supplementary, likelihood, gene, gaussian, equation, log, figure, average, global, data, approach, distribution] [row, clustering, column, matrix, planted, iterative, comparison, compute] [noise, cluster, three, bernoulli, estimated] [function, note, objective, problem, method, optimization, statistical, term, equality, size, arg, large, positive]
Causal Inference through a Witness Protection Program
Ricardo Silva, Robin Evans
Ricardo Silva, Robin Evans
[causal, set, ace, faithfulness, witness, admissible, provide, observational, wpp, graph, instrumental, procedure, variable, structure, binary, algebraic, adjustment, program, programming, interval, protection, polytope, patient, symbolic] [case, study, search, bound, notice, lower, algorithm, upper, extreme, context, conference, expected, probability] [table, pair, box, report] [model, posterior, figure, conditional, distribution, standard, approach, data, bayesian, sample, supplementary, full, prior, university, independence, parameter, estimate, generate, uncertainty, analytical, infer, true, average, chain, latent, inference, fully] [point, analysis, synthetic] [rule, system, common, observed, free, width] [linear, method, problem, discussed, assume, practice, function, theoretical]
Parallel Successive Convex Approximation for Nonsmooth Nonconvex Optimization
Meisam Razaviyayn, Mingyi Hong, Zhi-Quan Luo, Jong-Shi Pang
Meisam Razaviyayn, Mingyi Hong, Zhi-Quan Luo, Jong-Shi Pang
[selection, variable, set, consider, existing, number, subset, exists] [algorithm, result, lemma, implies, proof, decrease, order, serial, minimizing, best, general, bound] [proposed, propose, based, level] [parallel, approach, figure, university, processing, limit, separable, popular, stationary, supplementary] [approximation, arxiv, preprint, point, analysis, lim, numerical, minimization, nonconvex, greedy] [rule, behavior, simultaneously] [block, randomized, cyclic, iteration, coordinate, function, convex, bcd, descent, optimization, psca, convergence, update, objective, gradient, method, complexity, constant, problem, updated, nonsmooth, successive, large, desirable, lipschitz, solving, arg, theorem, practical, asymptotic, inequality, size, min, lasso, inf, assume, stochastic, xir, distributed, proximal, computational, communication, smooth, journal]
Incremental Clustering: The Case for Extra Clusters
Margareta Ackerman, Sanjoy Dasgupta
Margareta Ackerman, Sanjoy Dasgupta
[set, consider, ordering, structure, unique, core, store, random, number, fundamental, partition, typically, notion, list, discover, symposium, collection, subset] [algorithm, space, detect, learner, call, start, deterministic, lemma, setting, return, probability, minimum, allowed, turn] [additional, learning, select] [data, target, sequential, figure, larger] [clustering, incremental, nice, point, detected, distance, extra, analysis, niceness, balcan, recover, linkage, euclidean, ordered, small, weaker, clusterability, element] [cluster, perfect, distinct, state, time, memory, capture, single, include, induce] [batch, theorem, function, pick, cost, method, variant, suppose, positive, example, require, objective, class, replace, statistical]
Self-Paced Learning with Diversity
Lu Jiang, Deyu Meng, Shoou-I Yu, Zhenzhong Lan, Shiguang Shan, Alexander Hauptmann
Lu Jiang, Deyu Meng, Shoou-I Yu, Zhenzhong Lan, Shiguang Shan, Alexander Hauptmann
[map, number, increase] [easy, algorithm, best, action, called, setting, search, general, selecting, selects, starting] [spld, learning, spl, curriculum, training, diversity, selected, diverse, test, party, proposed, better, easiness, batchtrain, complex, table, select, indoorbirthday, olympic, learned, dev, outperforms, based, considering, gradually, birthday, video, baseline, climbing, med, minv, human, input, dataset, validation, reported, trained, conventional, train, indicates] [model, sample, approach, average, step, global, parameter, precision, figure, data] [comparison, clustering, spectral, denotes, robust] [event, three, represents] [term, regularization, loss, group, iteration, negative, method, iter, objective, threshold, problem, positive]
Hardness of parameter estimation in graphical models
Guy Bresler, David Gamarnik, Devavrat Shah
Guy Bresler, David Gamarnik, Devavrat Shah
[marginal, mapping, polytope, graphical, polynomial, graph, partition, approximating, set, fpras, marginals, exists, consider, paper, binary, structure, maximum, map, number, repulsive, random, characterization, procedure, ising] [lemma, proof, result, proposition, bounded, probability, bound, algorithm, goal, main, implies, determining, assuming] [learning, box, well] [model, backward, canonical, parameter, inference, exponential, variational, standard, conjugate, black, approach, approximate, forward] [vector, approximation, reduction, projected, corresponding, point, condition, pairwise, proved, requires, projection, project] [independent, property, time] [gradient, function, hardness, optimization, convex, problem, theorem, estimation, computational, showing, method, note, boundary, iterates, active, statistical, arg, strong, prove, complexity, class, basic, suppose, dual]
Expectation-Maximization for Learning Determinantal Point Processes
Jennifer A. Gillenwater, Alex Kulesza, Emily Fox, Ben Taskar
Jennifer A. Gillenwater, Alex Kulesza, Emily Fox, Ben Taskar
[set, runtime, marginal, variable, intelligence, constraint, consider, concave, recommendation, sum, naive, selection] [algorithm, conference, starting, probability, safety, simply, maximizing, setting] [learning, test, hidden, work, well, video, category, identity] [kernel, equation, determinantal, dpp, step, likelihood, log, baby, distribution, figure, process, develop, university, average, supplement, qyi, uncertainty, sampling, wishart, model, vyi, dpps, kulesza, feeding, data, bath, gear, approach] [matrix, point, projection, product, initialization, second, projected, eigenvectors, eigenvalue, eigendecomposition, corresponding] [ascent, derivative] [gradient, optimization, machine, update, size, relative, computational, negative, respect, psd, positive, example, note, simple]
Submodular Attribute Selection for Action Recognition in Video
Jingjing Zheng, Zhuolin Jiang, Rama Chellappa, Jonathon P. Phillips
Jingjing Zheng, Zhuolin Jiang, Rama Chellappa, Jonathon P. Phillips
[set, selection, subset, maximum, entropy, graph, edge, random, mixed, sum, number, consider, covered, vertex] [action, submodular, coverage, algorithm, selecting, initial, maximizing, histogram] [attribute, selected, recognition, discrimination, video, proposed, capability, based, discriminative, dataset, representation, walk, dda, hla, learning, table, propose, motion, accuracy, pair, human, olympic, performance, learned, feature, outperforms, score, select, combination, differentiating, extraction, svm] [figure, latent, data] [pairwise, dictionary, corresponding, sparse, vector, noisy, denote] [three, weight, represent, presented, redundant, contribution] [method, class, function, rate, large, weighted, size, objective, note, problem]
Scale Adaptive Blind Deblurring
Haichao Zhang, Jianchao Yang
Haichao Zhang, Jianchao Yang
[set, structure, typically, radius, number, quality, penalty, edge] [space, adaptive, uniform, benchmark, case, previous] [image, scale, deblurring, blind, proposed, blur, blurry, zhong, level, natural, presence, original, performance, based, zhang, deblurred, performs, multiple, well, levin, camera, work, tai, motion, achieve, deconvolution, pyramid, dataset, additional, understanding, conventional] [kernel, figure, approach, gaussian, model, latent, log, observation, perform, straightforward] [small, recovered, robust, analysis, noisy, formulation, smaller, sharp, point, sparse, denotes, corresponding, additive] [noise, observed, contribution, signal, single, estimated, contribute] [estimation, method, large, error, theorem, function, min, empirical, problem, cost]
Shape and Illumination from Shading using the Generic Viewpoint Assumption
Daniel Zoran, Dilip Krishnan, Jose Bento, Bill Freeman
Daniel Zoran, Dilip Krishnan, Jose Bento, Bill Freeman
[paper, solution] [algorithm, conference, result, current, order] [gva, shading, illumination, image, light, computer, lighting, generic, depth, mit, viewpoint, vision, ieee, ground, dataset, rendered, pixel, barron, scene, work, object, pattern, truth, scale, integrability, geometry, multiple, slope, novel, input, genericity] [model, surface, log, local, figure, estimate, spherical, linearization, international, allows, linearized, prior, data, contour, albedo, volume, supplementary] [estimating, alternating, small, cube] [shape, estimated, observed, single, stable, intrinsic, unstable, three] [term, estimation, optimization, cost, method, solve, direction, error, function, admm, example, derive, framework, assumption, linear, quadratic, constant]
Stochastic variational inference for hidden Markov models
Nicholas Foti, Jason Xu, Dillon Laird, Emily Fox
Nicholas Foti, Jason Xu, Dillon Laird, Emily Fox
[number, set, consider, belief, factor] [algorithm, transition, setting, sequence, current, optimal] [natural, learning, entire, based, dataset, hidden, performance, human, long, segmentation, learn] [variational, local, svihmm, inference, subchain, data, global, subchains, chain, markov, bayesian, approximate, latent, hmm, distribution, hmms, step, svi, parameter, approach, growbuf, full, chromatin, model, splash, series, elbo, observation, minibatch, predictive, sampling, posterior, true, international, unbiased, scaling, buffering] [synthetic, approximation, noisy, small, leading, compare] [length, time, state, independent, memory, encode, redundant] [stochastic, batch, gradient, large, update, machine, computational, buffer, error, assumption, iteration, optimization, convergence]
Object Localization based on Structural SVM using Privileged Information
Jan Feyereisl, Suha Kwak, Jeany Son, Bohyung Han
Jan Feyereisl, Suha Kwak, Jeany Son, Bohyung Han
[number, set, map, constraint, cutting, procedure, ratio, apply] [space, algorithm, upper, max, bound, search, called] [privileged, training, object, learning, localization, visual, structured, feature, based, bounding, svm, learn, joint, surrogate, segmentation, additional, testing, output, detection, lupi, overlap, image, performance, original, novel, learned, box, better, bird, improve, subwindow, attribute, ssvms, trained, vladimir, propose, handle] [model, inference, prediction, standard, data, average, figure, parameter, employed, approach] [corresponding, alternating, required, vector, small, denotes, knowledge] [weight, presented] [function, loss, problem, method, framework, arg, optimization, structural, ssvm, large, plane, generalization, objective, solved]
Efficient Inference of Continuous Markov Random Fields with Polynomial Potentials
Shenlong Wang, Alex Schwing, Raquel Urtasun
Shenlong Wang, Alex Schwing, Raquel Urtasun
[polynomial, energy, continuous, degree, graphical, set, arbitrary, random, concave, foe, sum, belief, exists, number, cccp, consider, fact, procedure, maximum, program, discrete, formulated, provide, mesh, mrfs] [algorithm, general, lower, space, proposition, order, lemma] [image, well, reconstruction, shading, denoising, table, based, effectiveness] [inference, approach, perform, markov, supplementary, global, multivariate, latent, gaussian, propagation, model, figure] [decomposition, basis, vector, hessian, compare, material, siam] [time, shape, evolution, curve, system] [convex, function, problem, convergence, optimization, linear, method, solving, min, solve, convexity, large, note, positive, theorem, achieves]
Structure Regularization for Structured Prediction
Xu Sun
Xu Sun
[structure, existing, set, reduce, strength, arbitrary, apply, random, graphical] [benchmark, general, algorithm, version, study, expected, position] [structured, training, based, work, learning, proposed, accuracy, crf, feature, natural, language, tag, task, accelerate, test, better, compared, evaluation] [prediction, sample, model, prior, local, figure, processing, distribution, markov, data, full, proposal] [analysis, sparsity, faster, suggests] [weight, effectively, raw, experimental, activity] [regularization, method, size, function, structreg, generalization, theoretical, structural, convergence, complexity, risk, weightreg, weightavg, objective, assume, rate, perc, sgd, cost, gradient, achieves, simple, convex, including, term, regularizing, comparing, stochastic, framework, metric]
Clustering from Labels and Time-Varying Graphs
Shiau Hong Lim, Yudong Chen, Huan Xu
Shiau Hong Lim, Yudong Chen, Huan Xu
[label, graph, qij, wmle, lij, partition, pij, consider, edge, corollary, program, unique, partial, existing, solution, set, partially, ratio, community, uij, number, disjoint, equal] [probability, setting, general, case, algorithm, result, social, main, considered, optimal, max] [pair, based, labeled, yij, performance, work] [log, model, mle, standard, approach, true, distribution, markov, generated, observation, figure, data] [clustering, matrix, planted, special, spectral, recovers, pairwise, recover, nuclear] [weight, observed, cluster, independent, interaction] [theorem, function, problem, weighted, convex, suppose, theoretical, block, assume, stochastic, large, note, simple, corresponds, constant, form, linear, including, applying]
A Latent Source Model for Online Collaborative Filtering
Guy Bresler, George H. Chen, Devavrat Shah
Guy Bresler, George H. Chen, Devavrat Shah
[user, item, recommendation, likable, collaborative, rating, recommended, number, random, tlearn, recommending, recommend, pui, consider, reedy, movie, ollaborative, high, provide, yui, clustered, solution, thought, decaying, xut, egood, consumed] [algorithm, probability, online, exploration, type, bandit, proof, good, exp, lemma, expected, reward, satisfy, goal, main, initial, choosing, study, guarantee, space] [learning, performance, work, similarity, ieee, dense, neighborhood, learn, based, joint, despite] [model, log, latent, step, standard, source, allow, mixture, explore] [matrix, top, clustering, condition, revealed, analysis, cosine, product] [time, system, song, event, preference, popularity] [problem, theorem, theoretical, assume, constant, example, note]
Large-scale L-BFGS using MapReduce
Weizhu Chen, Zhenghao Wang, Jingren Zhou
Weizhu Chen, Zhenghao Wang, Jingren Zhou
[number, dot, procedure, calculation, recursion, centralized, billion, scalability, reduce, store, mathematical, hashing, hash, easily, increase] [algorithm, base, environment, conference, search] [scale, learning, original, training, based, feature, performance, proposed, work, compared, produce, traditional, report, table] [data, step, implementation, approach, demonstrate, computation, historical, perform, advantage, parameter, parallel, massive] [product, requires, vector, huge, matrix, equivalent] [memory, auc, scalar, involving, time, length, cluster] [machine, distributed, large, update, calculate, gradient, direction, problem, optimization, relative, linear, complexity, method, require, storage, cost, updating, iteration]
Discrete Graph Hashing
Wei Liu, Cun Mu, Sanjiv Kumar, Shih-Fu Chang
Wei Liu, Cun Mu, Sanjiv Kumar, Shih-Fu Chang
[hashing, hash, discrete, graph, binary, lookup, hamming, code, dgh, set, maximization, bre, majorization, lsh, procedure, solution, itq, imh, sgn, klsh, isoh, nir, constraint, number, radius] [algorithm, search, max, space, optimal, good, initial, lemma] [training, learning, proposed, dataset, based, achieve, unsupervised, test, youtube, similarity, neighborhood, image, performance, object, face, learn, propose, higher] [data, local, sampled, tractable] [matrix, anchor, point, alternating, compact, distance, success, spectral, iterative, vector, small, denotes, randomly] [time, recall, directly] [problem, optimization, method, function, objective, query, large, linear, rate, convergence, solving, note]
Scalable Methods for Nonnegative Matrix Factorizations of Near-separable Tall-and-skinny Matrices
Austin R. Benson, Jason D. Lee, Bartek Rajwa, David F. Gleich
Austin R. Benson, Jason D. Lee, Bartek Rajwa, David F. Gleich
[mapreduce, set, structure, motivated, random, factor] [extreme, algorithm, choose] [transfer, selected, work, test, computed, geometry, based] [data, figure, separable, gaussian, simulation, plot, scalable, approach, advantage, larger, equation, university] [matrix, dimension, nonnegative, reduction, factorization, nmf, column, computing, separability, compute, technique, orthogonal, heat, xray, svd, residual, separation, projection, indexed, read, small, thin, spa, rank, tsqr, bubble, singular, hull, diagonal, analysis, unmixing, alternating, second, marker, conic, siam] [hyperspectral, cytometry, single, memory, three] [convex, require, relative, arg, error, large, note, block, problem, method, distributed, geometric]
Recovery of Coherent Data via Low-Rank Dictionary Pursuit
Guangcan Liu, Ping Li
Guangcan Liu, Ping Li
[number, paper, existing, solution, consider, random, construct, set] [algorithm, space, explored, conference, case] [computer, ieee, motion, performance, learning, unsupervised, pattern, vision, segmentation, handle, well] [data, figure, parameter, estimate, generated, international, full, log, model] [matrix, rpca, coherence, dictionary, lrr, rank, clustering, coherent, recovery, subspace, robust, sparse, column, corrupted, singular, exact, extra, denotes, randomly, analysis, row, norm, second, component, recovering, recover, emmanuel, guangcan, exactly, factorization, avoid, svd, successful, varies, pua, supported, principal, ssc, success, nonzero, gross] [cluster, underlying, produced, three] [error, problem, method, theorem, journal, machine, denoted, size, min, convex]
Feedforward Learning of Mixture Models
Matthew Lawlor, Steven W. Zucker
Matthew Lawlor, Steven W. Zucker
[number, multiview, discrete, set, variable, provide] [expected, general, access] [learning, input, learn, selective, presence, natural, hidden, multiple, sliding, work, indicates, based, visual] [mixture, data, model, converge, distribution, latent, processing] [tensor, decomposition, vector, formulation, denote, suggests, matrix] [rule, bcm, spike, synaptic, triplet, presynaptic, timing, independent, time, stable, classical, postsynaptic, voltage, drawn, neuron, neural, selectivity, hebbian, stimulus, poisson, spiking, occur, vpost, plasticity, stdp, ascent, noise, post, froemke, change, dan, free] [update, function, objective, stochastic, rate, gradient, assumption, note, assume, positive, negative, size, require, class, statistical, theory]
Analog Memories in a Balanced Rate-Based Network of E-I Neurons
Dylan Festa, Guillaume Hennequin, Mate Lengyel
Dylan Festa, Guillaume Hennequin, Mate Lengyel
[set, random, binary, balanced, high] [initial, gain, deviation, upper, satisfy, feedback] [network, pattern, performance, baseline, learning, visual, work, well, cat] [target, figure, distribution, model, parameter] [condition, distance, robust, point, small, spectral, successful, matrix, reduction] [memory, recall, inhibitory, activity, excitatory, synaptic, neuron, state, stable, neural, stability, noise, cortical, attractor, excitation, desired, graded, wij, weight, ssa, recurrent, time, observed, analog, ongoing, single, neuronal, dynamic, exc, connectivity, system, variability, inhibition, vinh, autoassociative, stimulus, trial, control, evolution, realistic] [rate, function, storage, large, note, ideal, computational, example, term]
Clustered factor analysis of multineuronal spike data
Lars Buesing, Timothy A. Machado, John P. Cunningham, Liam Paninski
Lars Buesing, Timothy A. Machado, John P. Cunningham, Liam Paninski
[factor, structure, assignment, variable, number, set] [initial, phase, algorithm, case, michael] [well, based, ground, applied, proposed, truth, reconstruction, learning, input] [model, data, latent, variational, inference, mixture, posterior, approach, uncertainty, prior, standard, parameter, estimate, observation] [clustering, subspace, matrix, analysis, dimension, sparse, spectral, union, component, approximation] [mixplds, cluster, activity, poisson, neural, loading, neuron, sorted, dynamical, preferred, recorded, liam, vem, motor, connectivity, observed, calcium, reasonable, spinal, time, population, spike, capturing, temporal, spiking, imaging, lds, cord, simultaneously] [group, method, linear, problem, large, estimation, function]
A Bayesian model for identifying hierarchically organised states in neural population activity
Patrick Putzky, Florian Franzen, Giacomo Bassetto, Jakob H. Macke
Patrick Putzky, Florian Franzen, Giacomo Bassetto, Jakob H. Macke
[tree, structure, number, binary, identify, node, discrete, set] [transition, probability, previous, best, decision, depend] [hidden, performance, well, multiple, understanding, evaluated, input, level, better] [model, data, variational, posterior, markov, inference, parameter, latent, distribution, bayesian, processing, regression, sampled, likelihood] [comparison, vary] [population, neural, cortical, state, estimated, stimulus, hierarchical, time, glm, activity, spiking, tuning, external, represent, generalised, hierarchically, emission, organised, modelled, recording, simulated, spike, organisation, sensory, temporal, decoded, decoding, gate, modulation, brain, presented, synchronised] [rate, statistical, linear, empirical, method, example, logistic, form, direction]
Scalable Kernel Methods via Doubly Stochastic Gradients
Bo Dai, Bo Xie, Niao He, Yingyu Liang, Anant Raj, Maria-Florina F. Balcan, Le Song
Bo Dai, Bo Xie, Niao He, Yingyu Liang, Anant Raj, Maria-Florina F. Balcan, Le Song
[random, number, exists, making, concept, high, preprocessing] [algorithm, optimal, bound, space, result, general, associated, order, best, previous, explicit] [training, learning, test, net, feature, achieve, scale, datasets, better, mnist, imagenet, dataset, based, novel, representation] [kernel, data, approach, figure, sample, computation, allows, university, gaussian, full, scalable] [approximation, matrix, support, analysis, hilbert, vector, small, rank, technical, fast, second] [neural, time, nonlinear] [stochastic, functional, function, error, gradient, method, doubly, rkhs, generalization, achieves, descent, theorem, rate, optimization, convex, sgd, storage, convergence, rtrue, cost, note, key, dual, coordinate, loss, theoretical, machine, measure, linear, norma, iteration]
Kernel Mean Estimation via Spectral Filtering
Krikamol Muandet, Bharath Sriperumbudur, Bernhard Schölkopf
Krikamol Muandet, Bharath Sriperumbudur, Bernhard Schölkopf
[solution, set, construct, consider, truncated, admissible, paper, jmlr] [space, probability, best, unknown, result, proposition, interesting] [learning, proposed, improve, supervised, embedding, work, improvement, table, natural, performance] [kernel, approach, figure, model, sample, standard, distribution, prior, parameter, gaussian, covariance, density, allow, mixture] [spectral, operator, estimating, support, basis, corresponding, minimization, role, condition, knowledge] [choice, range, inverse, interpretation, presented, tsvd] [shrinkage, estimator, empirical, theorem, function, landweber, regularization, risk, class, depends, estimation, rate, tikhonov, convergence, theoretical, note, form, problem, smoothness, iterated, consistent, accelerated, reproducing, measure, boosting, arg, regularized, kme, constant, assume, gradient, onb, iteration, resultant]
Sensory Integration and Density Estimation
Joseph G. Makin, Philip N. Sabes
Joseph G. Makin, Philip N. Sabes
[set, number, random, marginal, apply] [criterion, proof, case, sense, general, uniform, reliability, satisfy, optimal, guarantee, result, optimality, position] [trained, training, hand, visual, multiple, boltzmann, joint, learning, geoffrey] [model, density, data, mixture, distribution, latent, generative, true, posterior, inference, gaussian, statistic, probabilistic, estimate, conditional, prior, process, restricted, david, generated] [vector, special, condition, match] [multisensory, unisensory, neural, integration, stimulus, bernoulli, noise, observed, underlying, sensory, encoding, population, poisson, harmonium, property, broad, brain, weight, derived, corrupting, connection, mutual, alexandre, single, peak] [estimation, function, class, suppose, example, inequality, estimator]
General Table Completion using a Bayesian Nonparametric Model
Isabel Valera, Zoubin Ghahramani
Isabel Valera, Zoubin Ghahramani
[continuous, discrete, number, mixed, consider, provide, set, variable, linearly] [algorithm, general, conference, probability, case] [table, feature, proposed, dataset, database, object, based, attribute, propose, test, describe] [data, missing, gaussian, latent, ibp, observation, bpmf, model, probabilistic, sibp, gibp, approach, standard, bayesian, likelihood, categorical, auxiliary, distribution, figure, ordinal, inference, sample, average, supplementary, nonparametric, modeling, dealing, international, allows, processing, ynr, generated, indian, probit, nesarc, process] [matrix, completion, heterogeneous, vector, factorization, dimension, column, estimating, element] [real, count, wine, neural] [assume, weighting, positive, function, machine, note, denoted]
Dependent nonparametric trees for dynamic hierarchical clustering
Kumar Dubey, Qirong Ho, Sinead A. Williamson, Eric P. Xing
Kumar Dubey, Qirong Ho, Sinead A. Williamson, Eric P. Xing
[node, tree, assigned, number, structure, parent, set, marginal, root, random] [associated, online, algorithm, variation, probability] [learning, evaluation, performance, entire, proposed, indicates] [model, distribution, data, process, parameter, nonparametric, inference, figure, indian, allows, sample, gaussian, parallel, child, gibbs, stationary, covariate, mixture, modeling, sampling, markov, equation, concentration] [clustering, synthetic, vector] [time, dependent, topic, hierarchical, cluster, dirichlet, document, described, von, cold, vmf, tssbp, dtssbp, war, independent, modeled, chemistry, hierarchically, temporal, beta, three, conditioned, chinese, dynamic, evolving, change, stick, breaking, neural] [machine, generalized, journal, computational]
Sparse Bayesian structure learning with “dependent relevance determination” priors
Anqi Wu, Mijung Park, Oluwasanmi O. Koyejo, Jonathan W. Pillow
Anqi Wu, Mijung Park, Oluwasanmi O. Koyejo, Jonathan W. Pillow
[maximum, marginal, region, structure, number, selection] [relevance, result] [test, better, structured, based, automatic, multiple, learning, well, performance, input] [prior, regression, data, covariance, gaussian, bayesian, model, likelihood, determination, drd, figure, impose, fmri, local, posterior, inference, sdrd, estimate, approximate, log, distribution, csf, predictive, ard, process, hyperparameters, multivariate, variance, true, laplace, parameter, series, kernel, inducing] [sparsity, sparse, vector, matrix, fourier, diagonal, denote, approximation, ith] [simulated, weight, frequency, independent, dependent, estimated, noise, capture, underlying, real, distinct, response] [lasso, smoothness, method, group, smooth, error, estimation, large, strong, statistical, domain, squared, negative, machine, linear, size, optimization]
Mondrian Forests: Efficient Online Random Forests
Balaji Lakshminarayanan, Daniel M. Roy, Yee Whye Teh
Balaji Lakshminarayanan, Daniel M. Roy, Yee Whye Teh
[mondrian, tree, node, random, set, leaf, label, existing, forest, candidate, rab, ert, rabc, number, samplemondrianblock, structure, rectangle, root, quality, partition, roy, teh, cart] [decision, online, algorithm, associated, extend, probability, best, incrementally] [training, split, test, learning, accuracy, trained, dataset, represented, performance, achieve, location, additional, depth, work, labeled, ensemble] [data, distribution, sample, figure, model, predictive, bayesian, prediction, process, parameter, prior, evaluate, posterior, inference] [denote, point, dimension, introduce, incremental, refer, add, compare, discussion] [time, internal, independent, fraction, hierarchical] [batch, block, randomized, rate, complexity, machine, achieves, computational]
Parallel Sampling of HDPs using Sub-Cluster Splits
Jason Chang, John W. Fisher III
Jason Chang, John W. Fisher III
[number, set, ratio, switching, marginal] [algorithm, previous, deterministic, order] [split, propose, work, proposed, word, crf, joint] [sampling, global, local, figure, equation, sample, parallel, fsd, likelihood, gibbs, model, sampler, mjk, hastings, auxiliary, hdps, restricted, log, mcmc, zji, converge, mixture, distribution, markov, process, proposal, inference, data, developed, proposing, nytimes, posterior, chain, prior, processing, approximate, monte, dish, carlo] [detailed, approximation, synthetic] [topic, dirichlet, merge, hdp, single, inferred, merges, hierarchical, neural, expressed, directly] [convergence, method, journal, note, large, machine, augmented, converges]
Localized Data Fusion for Kernel k-Means Clustering with Application to Cancer Biology
Mehmet Gönen, Adam A. Margolin
Mehmet Gönen, Adam A. Margolin
[cancer, localized, genomic, fusion, dna, multiview, number, braf, nmi, rectal, colon, purity, set, clinical, sum, msi, performing, mutation, biomarkers, prognostic, mrna, methylation, patient, unweighted, atlas, solution, maximization, nen, zic, calculated, expression, assignment] [algorithm, maximize, conference, decision] [multiple, combination, learning, performance, better, feature, combine, similarity, propose, ieee, unsupervised, pattern, human] [kernel, data, approach, gene, qqq, figure, global, international, perform, processing] [clustering, matrix, copy, spectral, early, nonnegative, trace] [cluster, three, subject, neural, integration, nonlinear, biology] [problem, optimization, function, respect, solving, optimize, machine, note, solve, convex]
Learning with Fredholm Kernels
Qichao Que, Mikhail Belkin, Yusu Wang
Qichao Que, Mikhail Belkin, Yusu Wang
[set, consider, provide, number, ratio, marginal, high, paper] [space, optimal, setting, exp, case, call, alexander, choose] [learning, labeled, based, manifold, table, work, supervised, performance, mnist, proposed] [kernel, fredholm, data, unlabeled, target, gaussian, eqn, distribution, standard, density, integral, kpx, variance, equation, figure, prediction, bernhard, tsvm, laprlsc, rlsc, full, parameter, mikhail, exe, approach, varxe, international] [principal, approximation, noisy, low, synthetic, orthogonal, small, formulate] [noise, cluster, width, inverse] [linear, framework, problem, assumption, note, machine, loss, positive, theoretical, error, assume, inner, class, function, normalized, suppose, method, regularization, estimation, theorem, regularized, estimator]
Content-based recommendations with Poisson factorization
Prem K. Gopalan, Laurent Charlin, David Blei
Prem K. Gopalan, Laurent Charlin, David Blei
[user, collaborative, set, paper, complete, number, recommendation, consider, factor, variable, red, rating] [algorithm, expected, conference, draw] [word, better, computer, learning, based, natural] [latent, data, model, variational, inference, conditional, figure, likelihood, log, parameter, modeling, predictive, distribution, probabilistic, regression, auxiliary, international, bayesian, posterior, full, decoupled, gaussian] [factorization, matrix, arxiv, vector, top] [topic, ctpf, poisson, document, gamma, observed, ascent, rud, reader, article, wdv, ctr, shape, mendeley, represent, content, lda, conditionally, interpretable, multinomial, dirichlet, yud] [coordinate, rate, stochastic, statistical, machine, update, batch, including]
Reducing the Rank in Relational Factorization Models by Including Observable Patterns
Maximilian Nickel, Xueyan Jiang, Volker Tresp
Maximilian Nickel, Xueyan Jiang, Volker Tresp
[number, connected, runtime, variable, relation, partition, set, paper, existence, reduce] [lower, upper, observable, conference, lemma, bounded, social, algorithm, associated, oracle] [learning, performance, based, semantic, score, computed, proposed, evaluation, well, table] [data, model, latent, predictive, prediction, equation, approach, international, relationship, supplementary, figure, scalable] [rank, tensor, factorization, relational, rescal, adjacency, matrix, additive, diclique, digraph, required, multilinear, tucker, knowledge, link, material, recover, frontal, denotes, copy, rankpyk, component, kinship, decomposition] [single, include, observed] [theorem, large, complexity, term, machine, denoted, generalization, theoretical, size, note]
Conditional Random Field Autoencoders for Unsupervised Structured Prediction
Waleed Ammar, Chris Dyer, Noah A. Smith
Waleed Ammar, Chris Dyer, Noah A. Smith
[random, structure, induction, arbitrary, directed, maximum, undirected, number, runtime, quality, mrf] [sequence, appendix, exp] [crf, learning, unsupervised, word, autoencoder, autoencoders, feature, alignment, structured, discriminative, hidden, training, language, evaluation, translation, proposed, joint, supervised, work, natural, reconstruction, contrastive, learn, bitext, improve, brown, table, featurized, multiple, predicted] [model, inference, conditional, latent, hmm, generative, posterior, approach, markov, independence, local, data, standard, probabilistic, unlabeled, average, distribution, global, prediction, likelihood, dependency, series, inductive, blunsom, expectation] [requires, exact, syntactic] [observed, neural, emission, presented] [computational, machine, statistical, framework, effective, active, regularization, complexity, including]
Learning Generative Models with Visual Attention
Yichuan Tang, Nitish Srivastava, Ruslan R. Salakhutdinov
Yichuan Tang, Nitish Srivastava, Ruslan R. Salakhutdinov
[belief, set, energy, variable, number, region] [algorithm, good, interest, initial, best] [image, cmu, face, learning, training, attention, convnet, deep, gaze, caltech, visual, accuracy, learn, object, boltzmann, proposed, similarity, trained, hidden, dataset, based, iou, transformation, test, layer, gdbns, blue, computer, box, convolutional, train, pixel, propose, olshausen, geoffrey] [model, inference, generative, approximate, canonical, gaussian, hmc, step, dbn, data, distribution, average, posterior, modeling, conditional, latent, figure, gdbn, standard, sampling, big, log, monte, gibbs, hamiltonian, variational, cropped, local] [fast] [conditioned, neural, attentional, wij, visible, system] [function, note, framework, updated, large, gradient]
Sequence to Sequence Learning with Neural Networks
Ilya Sutskever, Oriol Vinyals, Quoc V. V. Le
Ilya Sutskever, Oriol Vinyals, Quoc V. V. Le
[map, set, number, john, google] [sequence, order, probability] [lstm, sentence, long, translation, input, training, bleu, learning, score, baseline, language, deep, beam, reversed, lstms, test, word, smt, performance, ensemble, rnn, work, network, english, rescoring, french, trained, reversing, task, garden, cho, card, achieved, dnns, well, output, mary, representation, hidden, qui, avec, une, rescore, sont] [source, model, target, approach, standard] [vector, close, arxiv, preprint] [neural, system, recurrent, short, length, state, vocabulary, single, time, memory, speech] [large, size, machine, problem, term, gradient, simple, method]
How transferable are features in deep neural networks?
Jason Yosinski, Jeff Clune, Yoshua Bengio, Hod Lipson
Jason Yosinski, Jeff Clune, Yoshua Bengio, Hod Lipson
[random, number, experiment, degree, red] [base, general, result, half, transition, upper, lower, call] [network, layer, trained, performance, dataset, transfer, imagenet, convolutional, training, task, transferred, better, natural, learning, deep, accuracy, selffer, split, learned, boost, feature, drop, anb, transferring, bnb, higher, datasets, transferability, baseb, dissimilar, train, work, learn, extent, color, well, visual, object, surprising, unsupervised, computer, image, separate, fragile, blue, jarrett, quantify, light, basea, generality, copied] [target, figure, treatment, university] [top, randomly, smaller, arxiv, second, compare] [neural, gabor, create, time, represent, exhibit, three] [generalization, large, splitting, optimization, error]
Design Principles of the Hippocampal Cognitive Map
Kimberly L. Stachenfeld, Matthew Botvinick, Samuel J. Gershman
Kimberly L. Stachenfeld, Matthew Botvinick, Samuel J. Gershman
[graph, map, set, structure, number, path, increase, discover] [grid, reinforcement, reward, transition, context, environment, policy, space, current, barrier, expected, discounted, maze, agent, probability, environmental] [place, learning, spatial, representation, work, location, segmentation, multiple, feature] [figure, estimate, track] [distance, eigenvectors, eigendecomposition, spectral, laplacian, eigenvector, matrix, second, analysis, decomposition, support] [hippocampal, cognitive, temporal, hierarchical, state, neural, detour, entorhinal, navigation, encode, brain, successor, cell, frequency, observed, hippocampus, memory, receptive, preexposure, annular, perceived, resemble, rewarded, european, expect, enclosure, encoding, facilitation, subgoals, simulated, cortex, fear, tolman, single, control, travel, response, population] [journal, function, direction, problem, normalized, rate, machine, computational]
Fast and Robust Least Squares Estimation in Corrupted Linear Models
Brian McWilliams, Gabriel Krummenacher, Mario Lucic, Joachim M. Buhmann
Brian McWilliams, Gabriel Krummenacher, Mario Lucic, Joachim M. Buhmann
[random, number, proportional, high, solution, typically, consider, reduce] [algorithm, setting, probability, difference, bound, michael, depend] [based, dataset, propose, additional, proposed, scale] [model, data, regression, bias, sampling, approximate, standard, observation, log, distribution, estimate, full, computation, figure, approach, variance, sampled, sample, relationship] [corrupted, leverage, ols, subsampling, approximation, matrix, nsubs, residual, subsampled, robust, knowledge, fast, uluru, srht, analysis, compute, small, suggests, point, reduction, equivalent, computing, projection] [described, simulated, noise] [randomized, weighted, error, estimation, statistical, linear, min, estimator, cost, term, large, delay, squared, arg, theorem, assumption, relative]
Sparse Space-Time Deconvolution for Calcium Image Analysis
Ferran Diego Andilla, Fred A. Hamprecht
Ferran Diego Andilla, Fred A. Hamprecht
[number, set, sum, quality, typically, maximum] [sequence, heuristic, upper] [spatial, image, learning, convolutional, deconvolution, representation, background, learned, convolution, accuracy, proposed, activation, appearance, coding, detection, train, work, pattern, well, location] [data, true, figure, multidimensional, automated, allows, prior, estimate, exponential, model] [sparse, fast, formulation, matrix, component, factorization, nonnegative, dictionary, product, analysis] [cell, time, impulse, activity, calcium, temporal, spike, response, single, imaging, sstd, neural, estimated, noise, std, real, sensitivity, filter, neuronal, cartesian, raw, signal, distinct, vivo] [optimization, large, method, block, problem, journal]
Spatio-temporal Representations of Uncertainty in Spiking Neural Networks
Cristina Savin, Sophie Denève
Cristina Savin, Sophie Denève
[random, potential, construct, linearly, number] [probability, main, depend] [network, representation, coding, input, idea, learning, level] [distribution, sampling, uncertainty, mcmc, model, probabilistic, computation, scheme, gaussian, posterior, target, allows, inference, highly, covariance] [condition, component, pairwise, matrix] [neural, decoding, underlying, spike, trajectory, encoding, time, decoded, stimulus, spiking, activity, tuning, neuron, recurrent, cortical, single, population, independent, membrane, variability, experimental, signal, ccg, represent, correlation, temporal, circuit, representing, fano, state, increased, measured, area, decoder, count, neuroscience] [distributed, linear, stochastic, example, computational, simple, function, size, class, consistent]
A Multiplicative Model for Learning Distributed Text-Based Attribute Representations
Ryan Kiros, Richard Zemel, Ruslan R. Salakhutdinov
Ryan Kiros, Richard Zemel, Ruslan R. Salakhutdinov
[consider, notion, correspond, set] [context, factored, interesting, illustrates] [word, language, attribute, sentence, learning, representation, english, blog, table, sentiment, network, learned, embedding, recursive, embeddings, author, text, atd, training, german, proposed, trained, learn, richard, computed, test, represented, work, performance, industry, performed, christopher, geoffrey, task, authorship, semantic, gender, metadata, french, train, evaluation, subphrases, jointly, andrew, character] [model, conditional, generate, parallel, allows, figure] [vector, tensor, matrix, denotes, compute, corresponding, denote, small] [neural, multiplicative, conditioned, corpus, three, capture, represent, inferred, document, correlation, recurrent, experimental] [distributed, method, linear, size, dimensional, framework, corresponds]
Deep Symmetry Networks
Robert Gens, Pedro M. Domingos
Robert Gens, Pedro M. Domingos
[map, set, number, hierarchy] [space, conference, lower, best] [symmetry, feature, deep, layer, symnet, object, pooled, symnets, transformation, translation, convnets, learning, image, convolutional, invariance, convnet, network, norb, net, computer, pattern, recognition, neighborhood, training, ieee, better, rotation, architecture, vision, generating, identity, test, dataset, datasets, pooling, visualized, arranged, represented, trained, rigid, representation, softmax, evaluation, computed] [kernel, figure, sample, model, larger] [point, vector, small, element, compute, square, person, matrix] [control, neural, represent, weight, underlying] [group, size, relative, large, machine, theory, generalization, locally, complexity, function, form]
A Residual Bootstrap for High-Dimensional Regression with Near Low-Rank Designs
Miles Lopes
Miles Lopes
[law, fact, high, consider, random, structure, approximating, paper, interval] [design, chosen, setting, result, bound, study, lemma, coverage, sequence, bounded, case, sense, focus] [work, based, well, testing, refers] [distribution, bootstrap, gaussian, regression, approximate, parameter, estimate, model, approximates, standard, variance] [approximation, ridge, matrix, vector, absolute, mspe, notation, condition, second, residual, bickel, sparsity, low, row, highdimensional, arxiv, point, preprint, decomposition, compute] [centered, conditionally, decay, width, response, drawn, choice] [theorem, estimator, linear, problem, method, suppose, empirical, consistency, simple, assume, error, note, normal, regularization, derive, assumption, annals, large, practical]
Sparse Polynomial Learning and Graph Sketching
Murat Kocaoglu, Karthikeyan Shanmugam, Alexandros G. Dimakis, Adam Klivans
Murat Kocaoglu, Karthikeyan Shanmugam, Alexandros G. Dimakis, Adam Klivans
[set, polynomial, random, unique, parity, hypergraph, graph, cut, number, program, user, maximum, hyperedges, runtime, high, provide, feasible, message, consider, linearly, messenger] [algorithm, probability, setting, relevant, uniform, unknown, lemma, opt, difference, chosen] [learning, dataset, learn, learns, pattern, work, labeled, based, ieee] [sample, approach, supplementary, distribution, data, log] [sparse, compressed, sensing, corresponding, boolean, vector, small, hypergraphs, sketching, fourier, hyperedge, matrix, randomly, learngraph, noisy, uniformly, analysis, denote, row, noiseless, exactly, amazon] [time, property, real, observed, drawn] [function, sign, problem, error, complexity, size, note, theorem, large, form, linear]
Convolutional Kernel Networks
Julien Mairal, Piotr Koniusz, Zaid Harchaoui, Cordelia Schmid
Julien Mairal, Piotr Koniusz, Zaid Harchaoui, Cordelia Schmid
[set, map, operation, number, consider] [space, algorithm, best, version, main] [convolutional, image, feature, layer, training, network, learning, patch, multilayer, visual, natural, recognition, invariant, input, deep, spatial, cnns, pooling, representation, unsupervised, learned, work, rpk, invariance, represented, table, convolution, better, performance, mnist, color, location, learn, downsampling, replacing, pixel, propose, built, build, accuracy, object] [kernel, data, gaussian, figure, approximate, sampling, approach, scheme, local, integral] [approximation, consists, hilbert, vector, small, exactly] [neural, shape, hierarchical, centered, representing] [linear, simple, gradient, size, method, function, normalized, regularization]
Learning Deep Features for Scene Recognition using Places Database
Bolei Zhou, Agata Lapedriza, Jianxiong Xiao, Antonio Torralba, Aude Oliva
Bolei Zhou, Agata Lapedriza, Jianxiong Xiao, Antonio Torralba, Aude Oliva
[set, number, high, forest, random, path, paper] [benchmark, current] [scene, sun, imagenet, database, diversity, training, image, datasets, performance, deep, feature, category, train, dataset, network, visual, recognition, test, accuracy, room, outdoor, trained, object, pool, attribute, cnn, pair, selected, convolutional, mit, architecture, garden, table, learning, human, visualization, learn, shop, computer, kitchen, layer, diverse, composed, generic, station, indoor, bedroom] [density, figure, data, average, field, larger] [randomly, compare, introduce, comparison] [neural, three, common, classification] [relative, measure, large, linear, nearest, require]
Generative Adversarial Nets
Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, Yoshua Bengio
Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, Yoshua Bengio
[adversarial, high, needed, procedure, maximum] [probability, algorithm, game, conference, criterion, maximize, minimax, minimum, optimal, order, space, case, previous] [training, deep, learning, work, trained, network, discriminative, train, input, net, backpropagation, multilayer, hidden, better, boltzmann, generating, table, improve, dropout, mnist] [generative, model, pdata, inference, data, distribution, log, discriminator, generator, approximate, markov, sample, likelihood, variational, modeling, global, approximated, density, parzen, conditional, figure, minibatch, international] [technical, second, point, minimization] [neural, noise, predictability, represent, competition, directly] [function, stochastic, framework, machine, estimation, gradient, theoretical, objective, require, optimization, convex, differentiable, update]
On the Number of Linear Regions of Deep Neural Networks
Guido F. Montufar, Razvan Pascanu, Kyunghyun Cho, Yoshua Bengio
Guido F. Montufar, Razvan Pascanu, Kyunghyun Cho, Yoshua Bengio
[number, set, maximal, map, consider, region, identify, subset, belief] [space, piecewise, exponentially, result, max, proposition, upper, universal, bounded, general, lower, bound, study, interesting, probability] [deep, hidden, layer, network, maxout, input, computed, shallow, activation, unit, computable, folding, hyperplane, hyperplanes, identifies, output, convolutional, rnl, based] [model, parameter, exponential, international, figure, restricted] [compute, corresponding, analysis, approximation] [neural, rectifier, feedforward, identified, pascanu, behavior, arrangement, replicated, intermediary, common, single, preactivation, choice, effectively, width] [linear, function, complexity, note, coordinate, theoretical, theorem, machine, example]
Optimistic Planning in Markov Decision Processes Using a Generative Model
Balázs Szörényi, Gunnar Kedenburg, Remi Munos
Balázs Szörényi, Gunnar Kedenburg, Remi Munos
[node, number, tree, set, high, label, structure, root, partial, path, consider, factor, smallest, parent, subtree, leaf] [policy, action, algorithm, planning, transition, bound, probability, order, optimistic, reward, case, optimal, called, mdp, decision, deterministic, upper, agent, uniform, search, previous, online, branching, secondary, bandit, initial, current, access, result, conference, expected, lower, reinforcement] [depth, labeled, work, based, better] [sample, child, generative, model, markov, approach, sampling, supplementary, log, global, estimate] [denote, analysis, close, sparse] [state, time, dynamic] [complexity, note, problem, stochastic, machine, depends, denoted, quantity, corresponds, theorem]
An Integer Polynomial Programming Based Framework for Lifted MAP Inference
Somdeb Sarkhel, Deepak Venugopal, Parag Singla, Vibhav G. Gogate
Somdeb Sarkhel, Deepak Venugopal, Parag Singla, Vibhav G. Gogate
[map, integer, polynomial, number, programming, solution, set, singleton, variable, program, assignment, consider, partial, existing, disjoint] [algorithm, search, return, appears, space, upper, called] [ground, network, based, idea] [inference, markov, approach, probabilistic, true] [denote, faster, relational, smaller] [weight, time, three, share] [lifted, mln, ipp, problem, predicate, ilp, formula, logic, mlns, tuffy, atom, weighted, conditioning, domain, lmap, logical, argument, cost, sarkhel, theorem, denoted, propositional, example, size, normal, grounding, solve, decomposer, linear, key, function, isolated, objective, assume, aly, solver, statistical, solving, alchemy]
The limits of squared Euclidean distance regularization
Michal Derezinski, Manfred K. Warmuth
Michal Derezinski, Manfred K. Warmuth
[random, number, set, fact, grows, high, hard] [algorithm, bound, lower, result, version, probability, space, conference, general, type, counting, open] [learning, combination, training, learn, feature, labeled, predict, based, computer, unit, additional, natural, help, feed] [average, figure, target, prediction, plot] [matrix, vector, square, rank, distance, small, euclidean, dimension, hadamard, column, singular] [weight, neural, single, described, produced] [linear, loss, problem, hardness, gradient, class, expanded, descent, predicts, function, simple, conjecture, sign, regularization, large, squared, note, instance, logistic, family, theorem, expansion, exponentiated, hold, dimensional, generalization, margin, prove, confused]
Spectral Clustering of graphs with the Bethe Hessian
Alaa Saade, Florent Krzakala, Lenka Zdeborova
Alaa Saade, Florent Krzakala, Lenka Zdeborova
[bethe, number, graph, set, community, belief, random, solution, cin, structure, degree, paramagnetic, radius, informative, remains, energy, exists, equal, ising, continuous, label] [case, algorithm, transition, phase, detect, relevant, open, best] [overlap, well, performance, network, computed] [model, generated, approach, standard, density, propagation, analytically] [hessian, spectral, clustering, matrix, eigenvalue, operator, bulk, cout, symmetric, eigenvector, assortative, spectrum, point, corresponding, second, adjacency, disassortative, eigenvectors, sparse, smaller, knowledge, glass] [real, free, choice] [block, stochastic, negative, positive, large, group, regularizer, function, theoretical, machine, statistical, theory, weighted, instance, respect]
Permutation Diffusion Maps (PDM) with Application to the Image Association Problem in Computer Vision
Deepti Pachauri, Risi Kondor, Gautam Sargur, Vikas Singh
Deepti Pachauri, Risi Kondor, Gautam Sargur, Vikas Singh
[structure, permutation, graph, number, synchronization, set, procedure, matchings, binary, pipeline, adjustment, provide, connected, existing, sum, frob, preprocessing, easily, synchronizing] [sfm, diffusion, good, base, called] [image, association, matching, reconstruction, rotation, computer, motion, oji, pdm, representation, bundle, camera, oat, repetitive, unitary, work, scene, based, proposed, baseline, datasets, dense, duplicate, visual, keypoints] [figure, global, data, local, approach, step, university, estimate, distribution] [matrix, match, fourier, exactly, symmetric, vector, pairwise, laplacian, corresponding] [derived, described, wij, single, choice] [group, problem, large, weighted, form, function, loss, consistent, relative, cup, optimization, consistency, linear, key]
Low-Rank Time-Frequency Synthesis
Cédric Févotte, Matthieu Kowalski
Cédric Févotte, Matthieu Kowalski
[maximum, set, sum] [algorithm, considered, setting, order] [gcm, learning, joint, ieee, traditional, work, transform, proposed, train, based] [model, log, likelihood, latent, gaussian, generative, data, approach, variance, source, bayesian, processing, mmle, standard, parameter, wiener] [matrix, analysis, sparse, nmf, nonnegative, second, factorization, separation, column, dimension, decomposition, noisy, vector, spectral] [signal, time, speech, synthesis, tonal, transient, frequency, lrtfs, audio, stft, noise, dis, described, estimated, gabor, relies, presented, music, window, snr, denoting, enhancement, hkn, wtransient, spectrogram, short, magnitude, raw] [estimation, example, iteration, min, large, negative, divergence, size, journal]
A State-Space Model for Decoding Auditory Attentional Modulation from MEG in a Competing-Speaker Environment
Sahar Akram, Jonathan Z. Simon, Shihab A. Shamma, Behtash Babadi
Sahar Akram, Jonathan Z. Simon, Shihab A. Shamma, Behtash Babadi
[variable, consider, high] [exp, order, algorithm, environment, switch, probability, chosen] [attention, proposed, entire, based, performance, well, human, denoising, inspired] [data, model, log, figure, estimate, process, distribution, parameter, observation, prior, simulation, prediction, third, series] [corresponding, estimating, condition, component, second] [meg, speaker, attentional, auditory, speech, attending, state, decoding, estimated, decoder, trf, time, snr, temporal, listener, neural, modulation, subject, real, window, attended, simulated, single, trial, response, correlation, unattended, indicate, spectrotemporal, listening, male, bernoulli, length, recorded, magnetic, dynamic, simon, envelope, ding] [estimation, framework, problem, journal, respect, function, method, denoted, linear]
Efficient Structured Matrix Rank Minimization
Adams Wei Yu, Wanli Ma, Yaoliang Yu, Jaime Carbonell, Suvrit Sra
Adams Wei Yu, Wanli Ma, Yaoliang Yu, Jaime Carbonell, Suvrit Sra
[constraint, structure, solution, number, high, paper] [algorithm, order, result, search] [structured, represented, ieee, propose] [figure, conditional, approach, full, true, data, observation, local, process, parameter] [matrix, rank, minimization, spectral, compressed, gcg, gcgls, norm, sensing, hankel, nuclear, low, formulation, vector, singular, siam, dppa, padmm, compare, recovered, running, column, row, approximation, formulate, factorization, reformulation, analysis, trace] [system, signal, time, observed, noise] [linear, problem, method, gradient, generalized, solve, iteration, objective, loss, cost, size, realization, min, complexity, optimization, computational, large, function, convex, stochastic, solving, convergence, consistent, subproblem, admm, iterate]
Efficient Minimax Signal Detection on Graphs
Jing Qian, Venkatesh Saligrama
Jing Qian, Venkatesh Saligrama
[connected, lmit, graph, subgraph, lattice, consider, subgraphs, node, thick, random, scan, arbitrary, county, composite, collection, clm, lmi, glrt, disease, asymptotically, outbreak, existing, number, maxt, set] [alternative, general, lower, case, bound, associated, optimal, max, minimax, upper, type, literature] [detection, test, performance, based, work, novel, hand, well] [gaussian, null, local, volume, computationally, standard, model, parameter, likelihood, log, separable, approach] [thin, analysis, matrix, denote] [shape, conductance, internal, signal, auc, poisson, connectivity, independent] [hypothesis, size, statistical, problem, convex, agnostic, family, asymptotic, linear, note, objective, large, normal, class, achieves]
Signal Aggregate Constraints in Additive Factorial HMMs, with Application to Energy Disaggregation
Mingjun Zhong, Nigel Goddard, Charles Sutton
Mingjun Zhong, Nigel Goddard, Charles Sutton
[energy, constraint, program, number, relaxed, paper, set, map] [total, conference, unknown] [hidden, learning, training, table, computed, applied, test, work, blind, proposed, entire] [data, disaggregation, aggregate, appliance, consumption, afhmm, afamap, model, chain, markov, posterior, nde, distribution, sit, factorial, log, sae, electricity, xit, source, household, prior, inference, load, equation, national, sitk, true, hmms, zit, figure, day, series, estimate, sac, average, incorporate] [component, vector, additive, denote, hit, computing] [signal, time, assumed, represent, estimated, measured] [problem, error, convex, quadratic, normalized, regularization, machine, size, objective, solve, optimization, computational, function, domain]
The Blinded Bandit: Learning with Adaptive Feedback
Ofer Dekel, Elad Hazan, Tomer Koren
Ofer Dekel, Elad Hazan, Tomer Koren
[set, switching, adversarial, edge, consider, random, number, arbitrary] [bandit, algorithm, player, regret, feedback, setting, blinded, online, round, sequence, switch, arm, action, lemma, oblivious, game, environment, adaptive, bound, eometric, chosen, observe, incurs, chooses, optimal, observes, cumulative, previous, called, draw, version, total, linded, best, focus, multiarmed, choose, probability, temporarily, result, extend, simply, conference, general, easy, exploration, auer, upper, uniform, satisfy, study] [learning] [standard, log, prediction, distribution, international] [revealed, vector] [time, measured, change, observed, independent, rule, presented] [loss, problem, prove, linear, theorem, optimization, update, rate, inequality, practical]
Near-optimal sample compression for nearest neighbors
Lee-Ad Gottlieb, Aryeh Kontorovitch, Pinhas Nisnevitch
Lee-Ad Gottlieb, Aryeh Kontorovitch, Pinhas Nisnevitch
[set, label, cover, solution, factor, corollary, covered, edge, subset, paper, interpoint, strictly, minimal, labeling, witness, covering] [algorithm, heuristic, minimum, bound, space, diameter, total, optimal, result, implies] [learning, compression, labeled, net, performance, evaluation, ieee, pattern] [sample, log, approximate, data, figure] [point, distance, doubling, dimension, included, wnnc, approximation, nnc, scaled, gadget, introduce, reduction, add, compute, covertype, improved, twin, low, consists, small] [weight, single, create, time, representing, property] [nearest, neighbor, problem, metric, consistent, margin, size, theorem, condensing, generalization, positive, negative, hardness, simple, instance, error, cost, machine, satisfying, assigning, empirical, function, assume]
Clamping Variables and Approximate Inference
Adrian Weller, Tony Jebara
Adrian Weller, Tony Jebara
[clamp, bethe, variable, attractive, partition, binary, clamping, zbi, graph, singleton, wmax, strength, random, maximum, tmax, maxw, coupling, consider, mpower, optimum, energy, topology, edge, belief, weller, dqi, marginal, easily, set, graphical, remaining, mrf, trw, vij, avg, summing, intelligence, clamped, fvs, qij, entropy, ratio, drj, increase, wainwright] [max, result, general, bound, best, worst, lower, upper, proof, observe, cycle, feedback, algorithm, good] [original, ieee] [log, model, approximate, average, figure, true, inference, local, propagation, series, variational] [approximation, pairwise, exact] [free, wij] [theorem, function, prove, arg, class, derive, convex, consistent, method, example, term]
A Block-Coordinate Descent Approach for Large-scale Sparse Inverse Covariance Estimation
Eran Treister, Javier S. Turek
Eran Treister, Javier S. Turek
[set, solution, number, random, expression, partitioning] [algorithm, chosen, appendix, choose] [table, dense] [covariance, approach, log, gene, full, data, gaussian] [matrix, sparse, small, minimization, condition, hessian, requires, synthetic, compute, analysis, iterative, clustering, computing, calculating, approximation, mentioned] [inverse, wij, memory, signal] [problem, block, method, newton, solving, gradient, direction, size, iteration, descent, solve, active, estimation, linear, quadratic, objective, activeij, updated, min, corresponds, linesearch, update, calculate, treating, positive, machine, computational, wnj, large, restricting, optimization, journal, lasso, cost, nlcg, statistical, term, note]
A Unified Semantic Embedding: Relating Taxonomies and Attributes
Sung Ju Hwang, Leonid Sigal
Sung Ju Hwang, Leonid Sigal
[set, label, number] [space, conference, main] [semantic, learning, embedding, category, discriminative, object, learned, learn, categorization, combination, embeddings, supercategory, plankton, attribute, longneck, uyi, vision, recognition, embed, visual, semantics, accuracy, based, learns, computer, work, ieee, pattern, proposed, transfer, multiclass, represented, cat, description, replacing, propose, correct, training, performance, improve, toughskin, table, novel, strainteeth, image, additional, describe, deep, test, ale, meatteeth] [model, auxiliary, parameter, generative, regression, data, generated, figure, international, precision] [sparse, multitask, decomposition, ridge, norm, knowledge, provided] [common, hierarchical, animal, neural, semantically] [class, regularization, method, objective, problem, machine, including, exclusive, optimization, large]
Predicting Useful Neighborhoods for Lazy Local Learning
Aron Yu, Kristen Grauman
Aron Yu, Kristen Grauman
[set, subset, label, apply, binary, number, candidate, existing, balanced] [good, best, relevant, space, ranking, algorithm] [training, learning, test, neighborhood, learn, attribute, predict, novel, visual, image, train, labeled, feature, yni, category, sun, dataset, xni, datasets, learned, trained, based, accuracy, table, baseline, jointly, apascal, automatically, similarity, selected, reconstruction, select] [local, data, global, approach, model, target, standard, estimate, generate, prediction, sample, series, regression] [vector, compressed, sensing, distance, support] [indicator, single, time, individual] [nearest, metric, neighbor, method, instance, size, function, weighted, lazy, active, example, domain, large]
Deep Fragment Embeddings for Bidirectional Image Sentence Mapping
Andrej Karpathy, Armand Joulin, Fei Fei F. Li
Andrej Karpathy, Armand Joulin, Fei Fei F. Li
[set, tree, relation, map] [ranking, previous, search, space, associated] [image, sentence, fragment, alignment, cnn, language, learning, score, visual, natural, network, multiple, detection, yij, word, embedding, work, evaluation, training, retrieval, imagenet, representation, socher, deep, similarity, convolutional, annotation, learns, test, multimodal, med, explicitly, task, describe, learn, performance, object, ground, complex, generating, semantic, entire, level, box, fullframe, ieee, truth] [model, global, dependency, figure, data] [corresponding, rank, top, vector, compute, skl, support, detected] [neural, triplet, described, common, single] [objective, method, positive, instance, consistent, example, note, machine]
Restricted Boltzmann machines modeling human choice
Takayuki Osogami, Makoto Otsuka
Takayuki Osogami, Makoto Otsuka
[set, strength, item, interval, binary, consider, increase, number, arbitrary, factor, degree] [probability, study, appendix, exp, decision, associated, choose, extend, observe, depend] [hidden, similarity, represented, train, training, boltzmann, dataset, selected, work, trained, car, learning, layer, adding, test, achieved, human, additional] [model, figure, restricted, prior, average, bias, distribution] [second, denote, consists, vector, corresponding, denotes, analysis] [choice, rbm, typical, represent, attraction, mlm, visible, compromise, phenomenon, independent, qxc, change, formally, keeping, length, represents, maglev, three, behavior, neural, preference, cognitive, range, hierarchical, include, presented] [theorem, rate, machine, denoted, divergence, empirical]
Multiscale Fields of Patterns
Pedro Felzenszwalb, John G. Oberlin
Pedro Felzenszwalb, John G. Oberlin
[binary, energy, random, set, consider, map, potential, marginal, number] [depend, algorithm] [image, multiscale, pattern, representation, training, pyramid, segmentation, work, baseline, detection, learning, natural, object, trained, pixel, scale, efop, hidden, better, salient] [model, fop, contour, markov, local, inference, figure, sampling, band, data, sampler, sample, generate, observation, approach, singlescale, distribution, coarsening, chain, prior, precision, posterior, estimate, mcmc, standard] [involves, small, vector, detecting, completion] [capture, single, described, time] [cost, block, function, update, stochastic, large, problem, framework, note]
Recursive Context Propagation Network for Semantic Scene Labeling
Abhishek Sharma, Oncel Tuzel, Ming-Yu Liu
Abhishek Sharma, Oncel Tuzel, Ming-Yu Liu
[labeling, tree, mapping, label, recursively, structure, node, random, map, hierarchy] [context, contextual, space, order, algorithm] [semantic, image, network, feature, rcpn, parse, recursive, scene, proposed, entire, training, pixel, ieee, cnn, stanford, performance, architecture, background, visual, better, accuracy, extraction, input, description, flow, based, crf, sift, learned, segmentation, trained, convolutional, parsing, dataset, structured, gpu, propagate, learning, table, output, scale, utilize, segment, learn, work] [local, model, propagation, figure, inference, global, prediction, water] [fast, smaller] [neural, time, individual, three, single, raw] [method, large, plain, size]
Encoding High Dimensional Local Features by Sparse Coding Based Fisher Vectors
Lingqiao Liu, Chunhua Shen, Lei Wang, Anton van den Hengel, Chao Wang
Lingqiao Liu, Chunhua Shen, Lei Wang, Anton van den Hengel, Chao Wang
[number, high, partition, increase, needed, set, apply] [lower, space, simply, result, best, general] [coding, fisher, image, feature, proposed, performance, scfvc, cnn, deep, table, traditional, gmmfvc, based, ieee, regional, outperforms, higher, activation, object, visual, achieve, achieved, scale, resolution, convolutional, representation, pascal, evaluation, extracted, accuracy, indoor, computer, layer, voc] [local, gaussian, model, gmm, distribution, kernel, process, global, figure, generative, comparable, average, mixture, normalization, demonstrated, data, gaussians] [vector, sparse, comparison, compare] [dimensionality, encoding, single, generation, neural, directly, three, choice, drawn, experimental] [dimensional, method, augmented, achieves, gradient, increasing, term, denoted, linear]
Weakly-supervised Discovery of Visual Pattern Configurations
Hyun Oh Song, Yong Jae Lee, Stefanie Jegelka, Trevor Darrell
Hyun Oh Song, Yong Jae Lee, Stefanie Jegelka, Trevor Darrell
[hard, set, graph, region, edge, informative, discovery, candidate, covering, identify, overlapping, cover, degree, mining, high, selection] [submodular, maximizing, intersection, maximize] [object, detection, discovered, discriminative, image, foreground, visual, learning, overlap, box, pascal, training, multiple, patch, weakly, supervised, automatically, bounding, trained, frequently, work, unsupervised, voc, test, select, challenging, localization, detector, adding, transform, selected, train, feature, better, performance] [figure, full, approach, data, generate, latent, estimate, lead] [frequent, approximation, greedy] [occur, single] [negative, positive, problem, relative, method, example, function, constrained]
Learning From Weakly Supervised Data by The Expectation Loss SVM (e-SVM) algorithm
Jun Zhu, Junhua Mao, Alan L. Yuille
Jun Zhu, Junhua Mao, Alan L. Yuille
[set, binary, candidate, label, map] [algorithm, gain, case] [object, segment, bounding, semantic, segmentation, performance, detection, pascal, positiveness, box, training, learning, groundtruth, level, voc, pixel, image, svm, supervised, svr, weakly, svc, train, feature, rcnn, evaluation, based, spatial, trained, supervision, imagenet, cpmc, iou, overlap, dataset, ubb, task, handle, computer, datasets, test, work, foreground, tapc, testing] [model, latent, proposal, prediction, data, average, expectation, standard, figure, approach] [support, vector, ndcg] [window] [method, problem, loss, function, regularization, positive, linear, class, term, weighted, threshold, instance, calculate, large, address, weak]
Self-Adaptable Templates for Feature Coding
Xavier Boix, Gemma Roig, Salomon Diether, Luc V. Gool
Xavier Boix, Gemma Roig, Salomon Diether, Luc V. Gool
[set, number] [algorithm, van, literature] [feature, input, pooling, quantization, image, learned, object, sift, accuracy, output, fisher, similarity, coding, based, training, performance, bsf, template, codebook, table, reported, resizing, report, computed, evaluating, terminology, recognition, vec, level, spd, representation, vision, computer, well, compared, spatial, descriptor, pascal, network] [average, standard, allows, equation] [vector, analyze, formulation, product, sparse, matrix, introduce, computing, tensor, denote, eigenvectors, square, second] [common, rest, logarithm, variability, hierarchical, dimensionality, analyzing] [note, function, inner, cost]
Elementary Estimators for Graphical Models
Eunho Yang, Aurelie C. Lozano, Pradeep K. Ravikumar
Eunho Yang, Aurelie C. Lozano, Pradeep K. Ravikumar
[graphical, set, discrete, random, consider, map, mapping, number, selection, graph, marginal, maximum, variable, structure, corollary, interior, marginals] [minimize] [well, table, learning, performance] [backward, parameter, log, model, mle, canonical, gaussian, exponential, covariance, true, btrw, markov, distribution, variational, approach, inference, sample, approximate, forward] [sparse, matrix, thresholding, pairwise, vector, approximation, corresponding, computing, norm, analysis, sparsity] [expressed] [regularized, statistical, class, estimator, family, problem, estimation, function, theorem, positive, optimization, convergence, simple, convex, derive, linear, nodewise, solve, large, suppose, regularization, proxy, note, key, quic]
Multivariate Regression with Calibration
Han Liu, Lie Wang, Tuo Zhao
Han Liu, Lie Wang, Tuo Zhao
[set, selection, solution, random, structure, consider, program, apply, high] [max, optimal, algorithm, oracle, design, lemma, appendix] [selected, performance, outperforms, based, table, human, learning, semantic] [regression, multivariate, parameter, prediction, model, log, figure, data, fmri, generated, gaussian, generate, series, standard, covariance] [matrix, norm, sparse, numerical, approximation, denote, frobenius, intermediate] [noise, activity, brain, tuning, adopt, stimulus, response] [cmr, omr, regularization, problem, statistical, achieves, argmin, function, admm, gradient, estimator, convergence, method, spg, linear, optimization, journal, loss, proximal, theorem, smooth, rate, estimation, solve, convex, assume, assumption, smoothed, objective, calibrates, smoothing, solved, machine, note, averaged, respect]
None
Deguang Kong, Ryohei Fujimaki, Ji Liu, Feiping Nie, Chris Ding
Deguang Kong, Ryohei Fujimaki, Ji Liu, Feiping Nie, Chris Ding
[solution, wgg, selection, set, structure, variable, number, uncorrelated, arbitrary, experiment, consider, overlapping] [algorithm, optimal, lemma, proof, correlated, mae, easy] [feature, learning, proposed, propose, selected, accuracy, dataset, datasets, compared, work, indicates, better] [data, standard, rmse, global, generate, log, figure, series] [sparsity, analysis, norm, matrix, synthetic, square, exactly, sparse] [correlation, indicate] [group, lasso, exclusive, plain, error, method, linear, problem, term, relieff, convex, function, loss, journal, regularization, solving, updating, solve, convergence, prove, machine, effective, statistical, ionosphere, gradient, note, isolet, leuml, theoretical, example, hub, rst, royal, fii, optimization]
Flexible Transfer Learning under Support and Model Shift
Xuezhi Wang, Jeff Schneider
Xuezhi Wang, Jeff Schneider
[number, selection, consider, apply, random, contained, set, sum, marginal, labeling, subset] [algorithm, case, ensure, result, goal, minimizes, optimal, best] [test, learning, transfer, labeled, training, work, proposed, better, transformation, dataset, selected, learn, adaptation, matching, transform, predict, propose, build, image, indicates, achieved] [data, target, model, teu, shift, tel, prediction, source, approach, grape, gaussian, kernel, conditional, wtel, covariate, yield, predictive, distribution, btel, covariance, figure, process, rtel, surveying, matched, kmm, allows, allow, rteu] [support, synthetic, corresponding, point] [offset, real, change] [active, domain, objective, method, problem, function, solve, smoothness, assumption]
Online combinatorial optimization with stochastic decision sets and adversarial losses
Gergely Neu, Michal Valko
Gergely Neu, Michal Valko
[set, consider, corollary, random, edge, independently, number] [regret, learner, bandit, decision, setting, leeping, policy, best, kanade, online, action, exploration, feedback, bound, algorithm, sleeping, bsfpl, combinatorial, proof, main, environment, availability, observe, case, total, space, phase, easy, associated, round, upper, omb, observes, sequence, bounded, neu, study, call, appendix, counting, guarantee, fpl, explicit, shortest, probability, considered] [learning, performance, work, based] [estimate, full, distribution, prediction, restricted, sequential, unbiased, observation, figure, standard, scheme] [vector, component, technique, estimating] [time, choice, interaction] [loss, problem, stochastic, theorem, optimization, method, min, assume, simple]
Distributed Balanced Clustering via Mapping Coresets
Mohammadhossein Bateni, Aditya Bhaskara, Silvio Lattanzi, Vahab Mirrokni
Mohammadhossein Bateni, Aditya Bhaskara, Silvio Lattanzi, Vahab Mirrokni
[mapping, coresets, balanced, coreset, set, mapreduce, number, graph, solution, increase, composable, map, partition, consider, notion, capacitated, bicriteria, rounding, quality, xuv, factor, concept, paper, nyc, wide, exists, radius, google, partitioning, applies, mergeable] [algorithm, space, main, total, general, proof, lemma, case, studied, upper] [based, table, work, input, original, idea] [model, sequential, data, computation, parallel, log, big, step, figure] [clustering, approximation, compute, streaming, union, denote, small, analysis, compressed, point] [cluster, memory, single, distinct, range] [distributed, problem, cost, size, theorem, framework, instance, constant, objective, prove, machine, linear, metric, suppose, function]