CS583 Probabilistic Graphical Models

Course Description

This course will cover probabilistic graphical models -- powerful and interpretable models for reasoning under uncertainty. The generic families of models such as directed, undirected, and factor graphs as well as specific representations such as hidden Markov models and conditional random fields will be discussed. The discussions will include both the theoretical aspects of representation, learning, and inference, and their applications in many interesting fields such as computer vision, natural language processing, computational biology, and medical diagnosis.

Course Topics

The following is a tentative and partial list of topics that will be covered in the class:

Bayesian networks
Markov networks
Factor graphs
Variable elimination
Belief propagation
Approximate inference
Structure learning
Hidden Markov models
Conditional random fields
Statistical relational learning

Course Information

Time and Location: Tue - Thu 3:15 - 4:30pm in Stuart Building 239
Professor: Mustafa Bilgic
Office: Stuart 228C
Email: mbilgic AT iit.edu
Office Hours: Tue - Thu 11am - 12pm (Other times by appointment)

Required background

Knowledge of probability and statistics is required. CS480 and CS584 are recommended but not required.

Course Format and Grading

In my couse, the slides often serve only as a guide; I use the white board heavily.

The evaluation will consit of written assignments, a project, a midterm, and a final. The point breakdown is:

Written assignments: 15%
Programming assignments: 15%
Midterm: 30%
Final: 40%

Late submission policy: You have seven free late days without penalty. You can use those days in any way you like for the written assignments and/or the project; you can use all of them for a single assignment or use them sparingly. However, once you run out of free days, late submissions will not be accepted. Absolutely positively no exceptions. So, use them wisely.
Collaboration policy: You can discuss the written assignments with your friends; however, everyone has to write their own solutions in their own words. You must include in your submissions the name of the people you collaborated with.
Code of academic honesty: Please read the procedures on academic honesty here. If you violate the academic honesty (such as unauthorized collaboration, cheating, etc.), then depending on the severity of the violation, it can result in i) getting zero points on the respective assignment, ii) expulsion from the course, iii) suspension of your enrollment at the university, iv) expulsion from the university.

Course Material

There is a required text book for this course:

Probabilistic Graphical Models, by Daphne Koller and Nir Friedman

There will be additional reading materials (mostly available on the web).

Tentative Schedule

Date	Topic	Reading
Jan 10	Syllabus & Introduction	Ch. 1
Jan 12	Foundations	Ch. 2
Jan 17	Bayesian networks	Ch. 3
Jan 19	Bayesian networks	Ch. 3
Jan 24	Markov networks	Ch. 4
Jan 26	Markov networks	Ch. 4
Jan 31	Local probabilistic models	Ch. 5
Feb 02	Local probabilistic models	Ch. 5
Feb 07	Template-based representations	Ch. 6
Feb 09	Variable Elimination	Ch. 9
Feb 14	Variable Elimination	Ch. 9
Feb 16	Clique Trees	Ch. 10
Feb 21	Clique Trees	Ch. 10
Feb 23	Approximate Inference	Ch. 12
Feb 28	REVIEW
Mar 01	MIDTERM EXAM
Mar 06	Approximate Inference	Ch. 12
Mar 08	MAP Inference	Ch. 13
Mar 13	Inference in temporal models	Ch. 15
Mar 15	Learning – overview	Ch. 16
Mar 20	SPRING BREAK
Mar 22	SPRING BREAK
Mar 27	Parameter estimation	Ch. 17
Mar 29	Parameter estimation	Ch. 17
Apr 03	Structure learning in Bayesian networks	Ch. 18
Apr 05	Structure learning in Bayesian networks	Ch. 18
Apr 10	Learning undirected models	Ch. 20
Apr 12	Collective classification
Apr 17	Collective classification
Apr 19	Hidden Markov models
Apr 24	Conditional random fields
Apr 26	REVIEW