I am a 1st year Computer Science Ph.D. student studying Machine Learning at Stanford University, advised by Percy Liang. I am grateful to be supported by the NDSEG Fellowship starting Fall 2019. Previously, I received my B.S. and M.S. in Computer Science in 2017, where I am grateful to have worked with Stefano Ermon on machine learning methods for sustainability, particularly in poverty mapping using satellite imagery.
Email: xie AT cs.stanford.edu
Robustness, interpretability, uncertainty, and extrapolation. Learning invariances and inductive biases, and encoding them with structure.
Transfer learning, domain adaptation, semi-supervised learning, unsupervised learning, meta-learning
Adversarial Training Can Hurt Generalization [Paper]
ICML Workshop on Identifying and Understanding Deep Learning Phenomena, 2019.
Reparameterizable Subset Sampling via Continuous Relaxations [Paper]
International Joint Conferences on Artificial Intelligence (IJCAI), 2019.Sang Michael Xie, Stefano Ermon
Semi-supervised Deep Kernel Learning: Regression with Unlabeled Data by Minimizing Predictive Variance. [Paper]
Neural Information Processing Systems (NeurIPS), 2018.
Association for the Advancement of Artificial Intelligence (AAAI), 2016. Oral Presentation, NVIDIA Global Impact Award Finalist, Scientific American 10 World Changing Ideas of 2016
Mapping Poverty with Satellite Imagery. [Paper]
Honors Thesis for B.S. with Honors
Semi-supervised Deep Kernel Learning. [Paper]
Neural Information Processing Systems (NIPS) Bayesian Deep Learning Workshop, 2016.
Incorporating Spatial Context and Fine-grained Detail from Satellite Imagery to Predict Poverty [Paper]
Jae Hyun Kim, Michael Xie, Neal Jean, Stefano Ermon
Semi-supervised Deep Kernel Learning. Amazon Graduate Symposium, 2019.
Transfer Learning from Deep Features for Remote Sensing and Poverty Mapping. Association for the Advancement of Artificial Intelligence (AAAI), 2016.