CIS 620 - Advanced Topics in AI - Spring 09
Probabilistic Graphical Models
Instructor: Ben Taskar Lectures: Monday and Wednesday, 10:30am-noon, Towne 319 Office hours: Monday, 1-3pm Courseschedule. Announcements:
Project guidelines
are posted. Project proposal is due on Wednesday,
March 4th.
Midterm is scheduled for Wednesday, March 18th, in class.
Final is on Monday, May 4th, 12:00-3:00.
Course description
Effective modeling of uncertainty and complexity of interactions is a fundamental problem in understanding and designing complex systems. By combining ideas from statistics and graph theory, probabilistic graphical models provide a general representation and an algorithmic framework for reasoning about statistical dependencies. The graphical model family includes very well-known members: Bayesian networks (BNs), Markov random fields (MRFs), hidden Markov models (HMMs), Kalman filters, Ising models, mixture models. These models underly many statistical approaches in computer vision, robotics, natural language understanding, signal processing, computational biology and other fields. The course will cover the following topics:
Part I: Representation
Markov properties
Directed graphical models
Undirected graphical models
Exponential family, generalized linear models
Part II: Inference
Elimination algorithm
Junction tree algorithm
Variational methods, mean field, belief propagation
Sampling methods, Gibbs, MCMC
Part III: Learning
Parameter estimation
Model selection
Incomplete data, expectation maximization (EM)
Mixture models, factor analysis, Dirichlet process
Evaluation
Grades will be based on a midterm (25%), project (25%) and final (50%).
Midterm (on 3/18/09) and final (on 5/4/09) are open book, open notes. More details about the project.
Materials
There is no published textbook for this course. A reader is available in the
copy room on the first floor of Levine. Photocopies of selected papers will be handed out in class.
Daphne Koller and Nir Friedman, Structured Probabilistic Models, in preparation.
Michael I. Jordan, An Introduction to Probabilistic Graphical Models, in preparation.
Pre-requisites
CIS 520 or equivalent. Knowledge of basic probabilty,
linear algebra, dynamic programming.