CS 228T: Advanced Topics in Probabilistic Graphical Models

This is an archive of materials used for CS 228T, taught at Stanford in 2011 with Daphne Koller.

Course description

An advanced course on probabilistic graphical models, covering advanced MCMC methods, variational inference, large margin methods, nonparametric Bayes, and other topics.

Prerequisites

The course requires CS 228 (probabilistic graphical models); CS 229 (machine learning) and EE 364A (convex optimization) are recommended.

Syllabus

• Advanced MCMC methods
• Variational inference
• MAP estimation
• Large margin methods
• Structure learning
• Latent variable models and topic models
• Bayesian nonparametrics

The syllabus may be adjusted through the course of the semester.

Notes

• Course summary and outline
• Convex optimization tutorial

Readings

The textbook is Koller and Friedman, Probabilistic Graphical Models, and various research papers will be assigned throughout the semester.

Homework

• Problem Set 1
• Problem Set 2
• Problem Set 3
• Problem Set 4

Quizzes

• Quiz 1
• Quiz 2
• Quiz 3
• Quiz 4
• Quiz 5
• Quiz 6
• Quiz 7
• Quiz 8