This class is part of the Master 2 MVA. It will last 18 hours (3x6 lectures) + 2h for the exam.
Final grade: approximately 70% final exam, 30% homeworks (to implement some of the algorithms seen in class). A single two-sided sheet of handwritten notes (with any content) will be allowed for the exam.
The homework is due by Friday, March 17, 2023. It is to be uploaded using the form here as a single pdf file of maximum size 40Mb. If the upload does not succeed (for some reason), send it by email to and but only after you have tried the upload.
It can be done alone or in groups of two students. The code can be done in any langage (python, R, matlab,...). The code file (e.g., notebook python, R) does not need to be delivered but the results and the figures must be included into the pdf report. The report can be written in English or in French.
Friday mornings from 9h00 to 12h00 at ENS Paris-Saclay. Information will be provided on the website before each class. Lecture notes will be updated here on the fly.
|1||13/01/2023||PG||1B18||Introduction. Sequential learning with individual sequences. Learning from expert advice.|
|2||20/01/2023||PG||Visio||Online convex optimization. Exponentiated Gradient. Online gradient descent.|
|4||03/02/2023||RD||1B18||Stochastic Bandits 1. Basic algorithms: Explore-Then-Commit, Upper Confidence Bound, ε-greedy.|
|5||10/02/2023||RD||1B18||Stochastic Bandits 2. Linear and continuous bandits.|
|6||17/02/2023||RD||1B18||Stochastic Bandits 3. Lower bounds. Best arm identification.|
|17/03/2022||1B18||Exam (from 10:00 to 12:00). It will be in person at ENS Paris-Saclay. Previous exams: 2020, 2021, 2022|