CS 228: Probabilistic Graphical ModelsStanford / Computer Science / Winter 2014-2015 |
Announcements
Important announcements will be posted here and on Piazza.
General Information
- Time/Location:
- Lectures: Tue/Thu 2:15-3:30pm, Gates B1, 3-4 units (01/05-03/13)
- Office Hours: See calendar
- Final Exam: The final will be 7pm Wednesday March 18, 2015. Location: Nvidia Aud and Huang 018.
Instructor: Stefano Ermon
- Course Assistants:
- Yuling Liu (yulingl@cs.stanford.edu )
- Bryan McCann (bmccann@stanford.edu )
- Billy Jun (billyjun@stanford.edu)
- Dorna Kashef Haghighi (dkashef@stanford.edu)
- Isaac Caswell (icaswell@stanford.edu)
- Gunaa Arumugam Veerapandian (avgunaa@stanford.edu)
- Louis Eugene(leugene@stanford.edu)
- Aditya Palnitkar (aditpal@stanford.edu)
Calendar: Click here for detailed information of all lectures, office hours, and due dates.
Contact: Please use Piazza for all questions related to lectures and coursework. For other issues, you can reach the course staff at cs228staff@gmail.com. For SCPD students, please email scpdsupport@stanford.edu or call 650-741-1542.
Coursework
- Course Description:
Probabilistic graphical models are a powerful framework for representing complex domains using probability distributions, with numerous applications in machine learning, computer vision, natural language processing and computational biology. Graphical models bring together graph theory and probability theory, and provide a flexible framework for modeling large collections of random variables with complex interactions. This course will provide a comprehensive survey of the topic, introducing the key formalisms and main techniques used to construct them, make predictions, and support decision-making under uncertainty.
The aim of this course is to develop the knowledge and skills necessary to design, implement and apply these models to solve real problems. The course will cover: (1) Bayesian networks, undirected graphical models and their temporal extensions; (2) exact and approximate inference methods; (3) estimation of the parameters and the structure of graphical models.
Prerequisites: Students are expected to have background in basic probability theory, statistics, programming, algorithm design and analysis.
Required Textbook: Probabilistic Graphical Models: Principles and Techniques by Daphne Koller and Nir Friedman. MIT Press.
- Further Readings:
Modeling and Reasoning with Bayesian networks by Adnan Darwiche.
Pattern Recognition and Machine Learning by Chris Bishop.
Machine Learning: a Probabilistic Perspective by Kevin P. Murphy.
- Grading Policy:
Homeworks (70%): There will be five homeworks with both written and programming parts. Each homework is centered around an application and will also deepen your understanding of the theoretical concepts. Homeworks will be posted on Piazza.
Final Exam (30%): TBA
Piazza: You will be awarded with up to 5% extra credit if you answer other students' questions in a substantial and helpful way.
- Assignments:
Written Assignments: Homeworks should be written up clearly and succinctly; you may lose points if your answers are unclear or unnecessarily complicated. You are encouraged to use LaTeX to writeup your homeworks (here is a template), but this is not a requirement.
Collaboration Policy and Honor Code: You are free to form study groups and discuss homeworks and projects. However, you must write up homeworks and code from scratch independently without referring to any notes from the joint session. You should not copy, refer to, or look at the solutions in preparing their answers from previous years' homeworks. It is an honor code violation to intentionally refer to a previous year's solutions, either official or written up by another student. Anybody violating the honor code will be referred to the Office of Judicial Affairs.
- Submission Instructions:
Regular Students: Regular (non-SCPD) students should submit all homeworks in hard copies in class. Please do not email your homework solutions to us.
SCPD Students: SCPD students should submit all homeworks in electronic format (PDF only). Your assignments must be emailed to the CAs at cs228staff@gmail.com.
Late Homework: Lateness of homeworks will be measured in terms of class periods (the time between two consecutive lectures is one class period). You have two grace periods which you can use at any time during the term without penalty. For a particular homework, you can use only one late period (i.e., submit late homework the following lecture). Once you run out your two late periods, homework will NOT be accepted. Each late homework should be clearly marked as "Late" on the first page.
Regrade Policy: You may submit a regrade request if you believe that the course staff made an objective error in grading. To do this, you must come in person to the owner CA in charge of the given homework. Any request submitted over email or on Piazza will be ignored. Remember that even if the grading seems harsh to you, the same rubric was used for everyone for fairness, so this is not sufficient justification for a regrade. If the regrade request is valid, the CA will add your request to the list, which will get processed.
Syllabus (tentative, periodically updated throughout the quarter)
Many thanks to David Sontag, Adnan Darwiche, Vibhav Gogate, and Tamir Hazan for sharing material used in slides and homeworks.
Week Date Topic Readings Assignments 1 Jan. 5-9 Introduction, Probability Theory, Bayesian Networks Chapters 1-3, Appendix A Homework 1 released. Due January 20 2 Jan. 12-16 Undirected models Chapter 4 3 Jan. 19-23 Exact Inference Chapter 9 Homework 2 released. Due February 3 4 Jan. 26-30 Message Passing Chapter 10 5 Feb. 2-6 Variational Inference Chapters 8,11
Graphical models, exponential families, and variational inference (Section 3)Homework 3 released. Due February 17 6 Feb. 9-13 More Variational Inference, Sampling Chapter 12 7 Feb. 16-20 MCMC, MAP inference Chapters 12, 13 Homework 4 released. Due February 26
8 Feb. 23-27 Learning Bayes Nets Chapters 16, 17, 19 Homework 5 released. Due March 10 9 Mar. 2-6 Structure Learning Chapters 18, 20 10 Mar. 9-13 Advanced topics and conclusions Other Resources
There are many software packages available that can greatly simplify the use of graphical models. Here are a few examples: