Introduction to Graphical Models

Guillaume Obozinski - Simon Lacoste-Julien - Francis Bach
Ecole des Ponts, ParisTech - INRIA/ENS - INRIA/ENS

Master recherche specialite "Mathematiques Appliquees",
Parcours M2 Mathematiques, Vision et Apprentissage (ENS Cachan), 1er semestre, 2015/2016


Please Register for the class (mandatory to access the homework).
 - Please fill out this form to register by Friday October 2nd NOON; extension to Wednesday Oct 14th 9am.

[Note that filling the form does not bind you to take the class for credit; it is just so that we can create an account for you to access the material on the moodle.]
This year, the class will be taught in English. All homework should be submitted in English.

Classes will take place on Wednesdays from 9am to 12pm at ENS Cachan, in Amphi Curie.

Moodle for the class.

Dates of classes

Below is the tentative schedule, with scribe notes from last year that might be updated for some lectures.

Date Lecturer Topics Corresponding chapters in class notes Scribe notes
September 30th
Guillaume Obozinski Introduction
Maximum likelihood
5
Slides (intro)
Slides ML
Huu Dien Khue Le, Robin Benesse
lecture1.pdf
lecture1.zip
October 7th
Guillaume Obozinski Linear regression
Logistic regression
Generative classification (Fisher discriminant)
--
K-means
EM
Gaussian mixtures
6, 7
Slides Regression
--
10, 11
Slides EM
Aymeric Reshef, Claire Vernade
lecture2.pdf
lecture2.zip
Marie d’Autume, Jean-Baptiste Alayrac
lecture3.pdf
lecture3.zip
October 14th
Simon Lacoste-Julien Graph theory
Directed graphical models
Undirected graphical models
2
Jaime Roquero, JieYing Wu
(updated 11/29)
lecture4.pdf
lecture4.zip
October 21st
Simon Lacoste-Julien Exponential families
Information theory
8, 19
scribbled notes
Thomas Belhalfaoui, Lénaïc Chizat
lecture5.pdf
lecture5.zip
October 28th
Francis Bach Gaussian variables
Factor analysis
13, 14
Lucas Plaetevoet, Ismael Belghiti
lecture6.pdf
lecture6.zip
November 4th
Simon Lacoste-Julien Sum-product algorithm
HMM
4, 12
scribbled notes
Pauline Luc, Mathieu Andreux
lecture7.pdf
lecture7.tex
November 11th

No lecture

November 18th
Guillaume Obozinski Approximate inference I
Sampling
Variational inference
21
Khalife Sammy, Maryan Morel
(new!) lecture8.pdf
lecture8.zip
November 25th
Guillaume Obozinski Approximate inference II
Sampling
Variational inference
21
Basile Clément, Nathan de Lara
(new!) lecture9.pdf
lecture9.tex
December 2nd
Simon Lacoste-Julien Bayesian methods
Model selection
5.1 and 5.3
scribbled notes
Gauthier Gidel, Lilian Besson
(new!) lecture10.pdf
lecture10.zip
December 16th
(Amphi Curie)
Final Exam
January 6th
Batiment Cournot (C102-103)
Project poster session


Homework (tentative schedule)

Homework 1, due October 24th, 2015 (on the Moodle): Homework | Data.

Homework 2, due November 11th, 2015 (on the Moodle): Homework | Data

Homework 3, due January 6th, 2016 (on the Moodle; we highly recommend you submit it in December though): Homework | Data


Projects

Project proposals

The final project allows a further understanding of certain aspects of the course. The following schedule has to be respected.

November Choose a project (one or two students per projects, preferably two)
Before 11/18 Send an email to the three teachers, in which all members of your team are cc'ed to request our agreement on your choice of team and project topic.
Before 12/09 Send a draft (1 page) + first results, on the Moodle.
On 2016/01/06 Poster session in Batiment Cournot (C102-103) - 9am to 12pm
Before 2016/01/13 Submit your project report (~6 pages, on the Moodle)


Description

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


References - Class notes

The course will be based on the book in preparation of Michael Jordan (UC Berkeley). Printed version of parts of the book (playing the role of the "polycopie") will be available from the Master's administrative assistant one or two weeks after the beginning of classes. We will notify you when they are ready for you to go and pick them up.


Last updated: December 2nd, 2015.