Introduction to Probabilistic Graphical Models and Deep Generative Models

Pierre Latouche - Pierre-Alexandre Mattei - Mathieu Even
UCA+Ecole Polytechnique+IUF+Inria

Master recherche specialité "Mathématiques Appliquées",
Parcours M2 Mathématiques, Vision et Apprentissage (ENS Paris-Saclay), 1er semestre, 2026/2027

Lectures will be given on Thursdays (9h-12h, 4 lectures) or Fridays (14h-17h, 5 lectures). This year, the students will be evaluated by a project [poster + notebook] on a research paper and a written exam. All lectures will be on zoom and recorded. The lectures at ENS Paris Saclay will be used to answer questions in person, regarding the lectures and the project. All documents (slides / blackboard / lecture notes) are or will be put here. The teaching assistant for this course is Rémi Khellaf.


Description

This course provides a unifying introduction to probabilistic modelling through the framework of graphical models, together with their associated learning and inference algorithms. It is an historical course at the core of the MVA program. Recent developments in mathematics are largely covered. In particular, this course cover deep latent variable models, variational autoencoders, energy based models, score-based diffusion models, amortised inference, and causal inference.

References

The course is based on the books Pattern Recognition and Machine Learning and Deep learning - Foundations and Concepts by C. Bishop. It is also based on the books Probabilistic Machine Learning: An Introduction and Probabilistic Machine Learning: Advanced Topics by K. Murphy. Finally, it relies on the book Causal Inference: A Statistical Learning Approach by Stefan Wager.

Projects

Internship offers

Dates of classes

Date Lecturer Topics
October 1st, Thursday, 9h-12h

ENS Paris Saclay + zoom
Pierre Latouche Introduction to directed graphical models
Maximum likelihood
Exemples (linear regression, logistic regression, neural networks)
October 8th, Thursday, 9h-12h

ENS Paris Saclay + zoom
Pierre Latouche K-means
EM
Gaussian mixtures
PPCA
October 15th, Thursday, 9h-12h

zoom
Pierre Latouche Bayesian linear regression
Gaussian processes
EM revisited
Model selection
October 22nd, Thursday, 9h-12h

ENS Paris Saclay + zoom
Pierre Latouche Approximate inference I: variational techniques
Stochastic block models + VEM
Expectation propagation
November 6th, Friday, 14h-17h

ENS Paris Saclay + zoom
Pierre-Alexandre Mattei Directed graphical models: theory and examples
November 13th, Friday, 14h-17h

zoom
Pierre-Alexandre Mattei Approximate Inference II: amortised variational inference
variational auto encoders
November 20th, Friday, 14h-17h

ENS Paris Saclay + zoom
Pierre-Alexandre Mattei Undirected graphical models, energy-based models, score-based diffusion models, approximate inference III: MCMC
November 27th, Friday, 14h-17h

ENS Paris Saclay + zoom
Mathieu Even Causal inference, potential outcomes, treatment effects and policy learning
December 4th, Friday, 14h-17h

ENS Paris Saclay + zoom
Pierre-Alexandre Mattei Deep generative models beyond likelihood-based models: GANs,
likelihood-free inference
December 17th, Thursday, 9h-12h

ENS Paris Saclay

Written exam


Last updated: July 2nd, 2026.