Probabilistic Graphical Models , Spring 2011

Probabilistic Graphical Models

Spring 2011

Overview

A graphical model is a probabilistic model, where the conditional dependencies between the random variables is specified via a graph. Graphical models provide a flexible framework for modeling large collection of variables with complex interactions, as evidenced by their wide domain of application, including for example machine learning, computer vision, speech and computational biology. This course will provide a comprehensive survey of learning and inference methods in graphical models, including variational methods, primal-dual methods and sampling techniques.

General information

Lecture: M-W-F 1:30-2:20pm
Room: TTI-C 5th floor

Instructor: Raquel Urtasun and Tamir Hazan
E -mail: rurtasun@ttic.edu, tamir@ttic.edu

Grading: exercises (50 %) + exam (50 %)

Book: Probabilistic graphical models: principles and techniques. Daphne Koller and Nir Friedman - MIT Press (2009)

EXAM: Friday June 10 at 10am. Will last for approximately 2.5h.

Syllabus

  1. Introduction:
    1. what's going to be covered in the class?
    2. introduction to probability and graphs
  2. Models
    1. Bayesian Networks
    2. Undirected Graphical Models
  3. Inference
    1. Exact Inference
    2. Sampling methods
    3. MAP inference
  4. Learning

Schedule

Lecture Date Topic

Slides

 Instructor Readings Assignments
1 March 28 Introduction lecture1.pdf Tamir Chapter 2

 
2 March 30 Bayesian Networks I lecture2.pdf Tamir Chapter 3

 
3 April 1 Bayesian Networks II lecture3.pdf Tamir Chapter 3

 
4 April 4 Undirected Graphical Models I
lecture4.pdf Raquel Chapter 4

 
5 April 6 NO CLASS
- - -

 
6 April 8 Undirected Graphical Models II
lecture5.pdf Raquel Chapter 4

 
7 April 11 Chordal Graphs, CRFs
lecture6.pdf Raquel Chapter 4

 
8 April 13 Exponential Family
lecture7.pdf Tamir Chapter 8 ex1 corrected typo! due April 20 at 1:30pm
9 April 15 NO CLASS: Snowbird
- - -

 
10 April 18 Exact inference I: VE
lecture8.pdf Raquel Chapter 9

 
11 April 20 Exact inference II: VE
lecture9.pdf Raquel Chapter 9

 
12 April 22 Exact inference III: Conditioning
lecture10.pdf Raquel Chapter 9

 
13 April 25 Exact inference IV: Clique Trees
lecture11.pdf Raquel Chapter 10

 
14 April 27 Exact inference V: Message passing I
lecture12.pdf Raquel Chapter 10

 
15 April 29 Exact inference VI: Message passing II
lecture13.pdf Raquel Chapter 10

 
16 May 2 Inference via optimization I
lecture14.pdf Raquel Chapter 11

 
17 May 4 Inference via optimization II
lecture15.pdf Tamir Chapter 11

 
18 May 6 Inference via optimization III
lecture16.pdf Tamir Chapter 11 ex2 due May 13 at 1:30pm
19 May 9 Inference via sampling I
lecture17.pdf Tamir Chapter 12

 
20 May 11 Inference via sampling II
lecture18.pdf Tamir Chapter 12

 
21 May 13 Inference via sampling III
lecture19.pdf Tamir Chapter 12

 
22 May 16 MAP estimation I
lecture20.pdf Tamir Chapter 13

 
23 May 18 MAP estimation II
lecture21.pdf Tamir Chapter 13

 
24 May 20 MAP estimation III
lecture22.pdf Tamir Chapter 13

 
25 May 23 Introduction to learning
lecture23.pdf Raquel Chapter 16 ex3 due May 30 at 1:30pm
26 May 25 Learning I
lecture24.pdf Tamir Notes

 
27 May 25 Learning II
lecture27.pdf Tamir Notes

 
28 May 27 Learning III
lecture27.pdf Tamir Notes

 
29 May 30 NO CLASS

 
30 June 1 Learning IV
lecture28.pdf Tamir Notes ex4 due June 7 at 1:30pm
31 June 3 Learning V
lecture29.pdf Tamir Notes