IFT 6269 : Probabilistic Graphical Models - Fall 2017

Description

This course provides a unifying introduction to statistical modeling of multidimensional data through the framework of probabilistic graphical models, together with their associated learning and inference algorithms.

People

Class info

Tentative content

(Detailed outline below)

Evaluation

Prerequisites

Textbook

Homework logistics

Detailed outline (updated often)

Below is a draft detailed outline that will be updated as the class goes on. For now it is the outline recopied from the Fall 2016 version of this class, with the links to the relevant old scribbled notes that will be updated with the new ones gradually. The related chapters in Mike's book are given (but note that they do not exactly correspond with the class content), and also sometimes pointers to the Koller and Friedman book (KF), the Deep Learning book (DL) or the Bishop's book (B). Related past ‘‘scribe notes’’ from the class that I taught in Paris are given for now, and will be updated with the scribe notes as I get them (if I get some).

Date Topics Related chapters
Scribbled notes
Scribe notes Homework milestones
Sept 5 Set-up & overview intro slides
lecture1
Isabela Albuquerque
lecture1.pdf
source
Sept 8 Probability review 2.1.1
DL: 3 (nice and gentle)
KF: 2.1 (more rigorous)
lecture2
William Léchelle (Fa16)
lecture2.pdf
source
Sept 12 Parametric models
Frequentist vs. Bayesian
5
lecture3
Philippe Brouillard and Tristan Deleu
lecture3.pdf
source
Hwk 1 out
Sept 15 Bayesian (cont.)
Maximum likelihood
lecture4 Philippe Brouillard and Tristan Deleu
lecture4.pdf
Sept 19 Statistical decision theory 1.3 in Bickel & Doksum
lecture5
Sébastien Lachapelle
lecture5.pdf
source
Sept 22 Properties of estimators
Linear regression
6
lecture6
Zakaria Soliman (Fa16)
lecture6.pdf
source
Sept 26 Linear regression (cont.)
Logistic regression
DL: 7.1 (l2, l1-reg.)
6, 7
lecture7
MVA lecture2 Hwk 1 due
Hwk 2 out
data.zip
Sept 29 Optimization
Gen. classification (Fisher)
Derivative tricks for Gaussian MLE
DL: 4.3
Boyd's book
Matrix Diff. book
lecture8
Oct 3 Kernel trick
K-means
Gaussian mixtures
EM
10, 11
lecture9
MVA lecture3
Oct 6 GMM and EM
Graph theory
10,11
2
lecture10
MVA lecture4
Oct 10
Today: room Z-305 McNicoll
Directed graphical models 2
lecture11
Hwk 2 due
Hwk 3 out
data.zip
Oct 13 DGM (cont.)
Undirected graphical models
2, 3
lecture12
Oct 17 Inference: elimination alg.
sum-product alg.
4
lecture13
MVA lecture7
Oct 20 Max-product
junction tree
HMM
17, 12
lecture14
Oct 24 Break: look at projects
Oct 27 Break: look at projects
Oct 31 HMM and EM
Information theory
12, 19
lecture15
MVA lecture5 Hwk 3 due
Hwk 4 out
Nov 3 Max entropy
Duality
19
lecture16
MVA lecture6
Nov 7 Exponential families

Sampling
8
lecture17
21
MVA lecture8 Project: team formed
Nov 10 Sampling (cont.) 21
lecture18
Nov 14 MCMC sampling 21
lecture19
Nov 17
(Sarath lecture)
Non-parametric models: Gaussian processes
Dirichlet processes
25
lecture20
DP slides
Nov 21 Gibbs sampling (cont.)
Variational methods
Bishop: 10.1
lecture21
Hwk 4 due
Hwk 5 out
Nov 24 Variational methods (cont.)
Estimation in graphical models
9
lecture22
Nov 28 Bayesian methods
Model selection
5, 26
lecture23
MVA lecture10 Project: 1 page
progress report due
Dec 1

Note: starting at 12:30pm
Gaussian networks

Factor analysis, PCA, CCA
(Kalman filter)
VAE
lecture24

13
old lecture17
14 , (15)
old lecture18
VAE – DL: 20.10.3
Dec 5 No lecture this week
work on your project!
Dec 12 Poster presentation
1:30pm-4:30pm
mezzanine of Jean-Coutu atrium
Hwk 5 due
Take-home final out
Dec 20 Project report due
Take-home final due