Probabilistic Reasoning in AI (COMP526)
Winter 2004
News

Partial solutions to homework 6 are now posted here.

Solutions to homework 5 are now posted here.

Homework 6 is now posted here. It
is due Thursday April 8, midnight.

Grades for the first midterm are now posted here.

Slides for lecture 18 have been posted Tuesday, March 23, 4:45pm.

An exam web page is available with the list of
topics, and possible dates.

The web page has been updated with materials for EM and temporal
models, lecture slides for lectures 17, 19, and homework 5 (due
Friday, March 26, in class).

The midterm tomorrow is in Leacock 26, 5:45pm. You are allowed
one doublesided cheat sheet. Partial homework solutions are now
posted here.

Homework 4 is now posted on the web page. It is due next Monday,
February 16. Note that homework 3 (programming junction tree
algorithm) will be posted tomorrow. Homework 4 covers material for
the exam, which is why it has been posted first.

Due to requests, I have posted my sketchy junction tree slides on the
web site. I strongly suggest that you look at them together with
chapter 17 of the textbook, which offers all the details.

All class materials have been updated. I have posted all lecture
notes and also materials from the textbook for the algorithms covered
so far. I also posted homework 2, due Friday, Feb. 6,
10pm

Slides for lecture 5 have been been updated on Friday, Jan. 16. The
placeholders have been replaced with actual slides. Also, the
textbook materials are now linked from the schedule web page
(restricted access).
 I have sent an email to my own class mailing list a couple of
days ago. It should have arrived at your mail.mcgill.ca account. If
you haven't got it, please send me an email so I can update my
address book.
General Information
Where: Burnside Hall, Room 1B24.
When: Monday, Wednesday, Friday, 1:352:25pm.
What: One of the primary goals of AI is the design, control
and analysis of agents or systems that behave appropriately in a
variety of circumstances. Good decision making often requires the
existence of knowledge or beliefs about the agent's environment, as
well as about its own abilities to observe and change the environment,
and about its own goals and preferences. In this course we will
examine computational approaches for modeling the environment and
solving decision problems. We will focus mainly on probabilistic
models of reasoning, and on sequential decision making.
The course is intended for advanced undergraduate students and for
graduate students, and will provide an introduction to the ongoing
research in the field of reasoning under uncertainty, which has been
very active during the last decade. We will cover the following
topics:
 The principles of Bayesian inference
 Belief networks
 Syntax and semantics
 Exact and approximate inference
 Learning methods

Temporal models
 Hidden Markov Models
 Dynamic Bayes nets
 Kalman filters
 Basics of utility theory
 Markov Decision Processes
 Dynamic programming methods
 Structured state and action spaces
 Temporaldifference learning algorithms
 Generalization and function approximation
 Partially Observable Markov Decision Processes
 Exact solution methods
 Approximate methods
Instructor
Doina Precup
School of Computer Science
Office: McConnell 326
Office Hours:
Wednesday, 2:304:00pm
Friday, 2:303:00pm.
Meetings at other times by appointment only
Phone: 3986443
Email: dprecup@cs.mcgill.ca
IMPORTANT: Email is the quickest way to
reach me and get your questions answered.
Teaching assistant
Rohan Shah
Office Hours: Monday and Friday, 45pm, McConnell 112. Meetings at other times by appointment only
Email: rshah3@cs.mcgill.ca
References
 Textbooks:
 Lecture notes and other relevant materials are available on this
web page.
Doina PRECUP
Last modified: Tue Apr 13 11:57:29 EDT 2004