Bayesian machine learning
Rémi Bardenet (CNRS, Univ. Lille) & Julyan Arbel (Inria, Univ. Grenoble-Alpes)
LearningMachine Learning

Prè-requis

  • An undergraduate course in probability.
  • It is recommended to have followed either the course of P. Latouche and N. Chopin on « Probabilistic graphical models » or the course of S. Allassonière on « Computational statistics » during the first semester.
  • Practical will be in Python and R. A basic knowledge of both languages is required. Nothing fancy, students should simply be able to read and write simple programs and load libraries: going through a basic online tutorial in both languages should be enough.

Objectif du cours

By the end of the course, the students should

  •  have a high-level view of the main approaches to making decisions under uncertainty.
  •  be able to detect when being Bayesian helps and why.
  •  be able to design and run a Bayesian ML pipeline for standard supervised or unsupervised
    learning.
  •  have a global view of the current limitations of Bayesian approaches and the research
    landscape.
  •  be able to understand the abstract of most Bayesian ML papers.

Organisation des séances

* 6×2 hours of lectures
* 4×2 hours of practicals
* 4 hours of « student seminars » for the evaluation.
* All classes and material will be in English. Students may write their final report either in French or English.

Mode de validation

  • Students form groups. Each group reads and reports on a research paper from a list. We strongly encourage a dash of creativity: students should identify a weak point, shortcoming or limitation of the paper, and try to push in that direction. This can mean extending a proof, implementing another feature, investigating different experiments, etc.
  • Deliverables are a small report and a short oral presentation in front of the class, in the form of a student seminar, which will take place during the last lecture.

Références

Parmigiani, G. and Inoue, L. 2009:Decision theory: principles and approaches. Wiley.

Robert, C. 2007. The Bayesian choice. Springer.
Murphy, K. 2012. Machine learning: a probabilistic perspective. MIT Press.
Ghosal, S., & Van der Vaart, A. W. 2017. Fundamentals of nonparametric Bayesian inference. Cambridge University Press.

Thèmes abordés

  • Decision theory
  • 50 shades of Bayes: Subjective and objective interpretations
  • Bayesian supervised and unsupervised learning
  • Bayesian computation for ML: Advanced Monte Carlo and variational methods
  • Bayesian nonparametrics
  • Bayesian methods for deep learning
voir les autres cours du 2nd semestre