Introduction to Graphical Models

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

Master recherche specialite "Mathematiques Appliquees",
Parcours M2 Mathematiques, Vision et Apprentissage (ENS Cachan), 1er semestre, 2014/2015




Please Register for the class (mandatory to access homeworks).
 - If you are registered at ENS Cachan, please go directly here
 - If not, please fill out this form. You will then get instructions by email.
This year, the class will be taught in English. All notes and assignment will be given in English. In class or for all assignments, students may use either French or English. 

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


Dates of classes

Note that some class notes (in French) are available for most of the classes from earlier years.
Date Lecturer Topics Corresponding chapters in class notes
 Scribe notes
October, 1st
Curie
 Guillaume Obozinski Introduction
Maximum likelihood
Linear regression		 5
Slides (intro)
Slides
 Huu Dien Khue Le, Robin Benesse
lecture1.pdf
lecture1.zip
October, 8th
Curie
 Guillaume Obozinski Logistic regression
Generative classification
K-means		 6, 7
Slides Regression
Slides EM
 Aymeric Reshef, Claire Vernade
lecture2.pdf
lecture2.zip
October 15th
Condorcet (Batiment d'Alembert)		 Francis Bach EM
Gaussian mixtures
Graph theory		 10, 11
 Marie d’Autume, Jean-Baptiste Alayrac
lecture3.pdf
lecture3.zip
October 22nd
Curie		 Francis Bach Directed graphical models
Undirected graphical models		 2
 Vincent Bodin, Thomas Moreau
lecture4.pdf
lecture4.zip
October 29th
Curie		 Guillaume Obozinski Exponential families
Mixture of experts		 8, 10, 19
 Thomas Belhalfaoui, Lénaïc Chizat
lecture5.pdf
lecture5.zip
November 5th
Curie		 Guillaume Obozinski Gaussian variables
Sum-product algorithm		 13, 14
 Lucas Plaetevoet, Ismael Belghiti
lecture6.pdf
lecture6.zip
November 12th
Curie		 Francis Bach HMM
Factor Analysis		 4, 12
 Pauline Luc, Mathieu Andreux
lecture7.pdf
lecture7.tex
November 19th
Curie		 Guillaume Obozinski Inexact inference 21
November 26th
No lecture

December 3rd
Curie		 Francis Bach Bayesian methods

Moussab Djerrab
lecture9.pdf
lecture9.tex



Homeworks

Homework 1, due October 22, 2014 (on the Moodle): Homework Data

Homework 2, due November 12, 2014 (on the Moodle): Homework Data

Homework 3, due January 7, 2015 (on the Moodle): 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.
Mid-november Choose a project (one or two students per projects, preferably two)
Before 11/26 Submit proposal detailing the choice of project. More information on the procedure will provided soon.
Before 12/03 Send a draft (1 page) + first results, on the Moodle.
12/17 Poster session in Batiment Cournot (C102-103) - 9am to 12pm
Before 01/07 Submit your project report (~6 pages, on the Moodle)
Before 01/07 Submit your take home exam (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) and on research articles.  The course notes ("polycopie") may be obtained from the Mastere's adminstrative assistant.