Rémy Degenne

About me

I am a researcher in the Scool Team-project at Inria centre at the University of Lille. I work mostly on bandit theory and I am interested in all forms of online and reinforcement learning as well as statistics and optimization.

I am a maintainer of the mathematical library mathlib for the Lean theorem prover.

In 2020 I was a post-doctoral researcher at Inria Paris, in the SIERRA Team. In 2018-2019 I spent a year in the Machine Learning group at CWI Amsterdam, working with Wouter M. Koolen. From 2015 to 2019 I was a PhD student under the supervision of Vianney Perchet at the CMLA research center of Ecole Normale Supérieure Paris-Saclay and the LPSM lab at Université Paris Cité.

News

[November 2023] I am contributing to the formalization of the polynomial Freiman-Ruzsa conjecture in Lean. This is a project led by Terence Tao, who recently proved this result with W. T. Gowers, Ben Green and Freddie Manners. See this post on Terence Tao's blog for a tour of the project.

[2022] My ANR JCJC project proposal FATE (Frugal and adaptive testing) was accepted.

[July 2022] I co-organize the Complex Feedback in Online Learning workshop at ICML 2022 (Baltimore, July 23). This workshop aims to present a broad overview of the feedback types being actively researched in sequential learning (reinforcement learning, bandits, games...), highlight recent advances and provide a networking forum for researchers and practitioners.

Publications

Last updated: December 2023.

• An ε-Best-Arm Identification Algorithm for Fixed-Confidence and Beyond, NeurIPS 2023,
  with Marc Jourdan and Emilie Kaufmann.
• Non-Asymptotic Analysis of a UCB-based Top Two Algorithm, NeurIPS 2023,
  with Marc Jourdan.
• Fast Asymptotically Optimal Algorithms for Non-Parametric Stochastic Bandits, NeurIPS 2023,
  with Dorian Baudry, Fabien Pesquerel and Odalric-Ambrym Maillard.
• On the Existence of a Complexity in Fixed Budget Bandit Identification, Colt 2023.
  [HTML Version]
  
• Dealing with Unknown Variances in Best-Arm Identification, ALT 2023,
  with Marc Jourdan and Emilie Kaufmann.
• A formalization of Doob's martingale convergence theorems in mathlib, CPP 2023,
  with Kexing Ying.
• Top Two Algorithms Revisited, NeurIPS 2022,
  with Marc Jourdan, Dorian Baudry, Rianne de Heide and Emilie Kaufmann.
• On Elimination Strategies for Bandit Fixed-Confidence Identification, NeurIPS 2022,
  with Andrea Tirinzoni.
• Choosing Answers in Epsilon-Best-Answer Identification for Linear Bandits, ICML 2022,
  with Marc Jourdan.
• Online Sign Identification: Minimization of the Number of Errors in Thresholding Bandits, NeurIPS 2021,
  with Reda Ouhamma, Vianney Perchet and Pierre Gaillard.
• Dealing With Misspecification In Fixed-Confidence Linear Top-m Identification, NeurIPS 2021,
  with Clémence Reda and Andrea Tirinzoni.
• Structure Adaptive Algorithms for Stochastic Bandits, ICML 2020,
  with Han Shao and Wouter M. Koolen.
• Gamification of Pure Exploration for Linear Bandits, ICML 2020,
  with Pierre Ménard, Xuedong Shang and Michal Valko.
• Non-asymptotic pure exploration by solving games, NeurIPS 2019,
  with Wouter M. Koolen and Pierre Ménard.
• Pure exploration with multiple correct answers, NeurIPS 2019,
  with Wouter M. Koolen.
• Bridging the gap between regret minimization and best arm identification, with application to A/B tests, AISTATS 2019,
  with Thomas Nedelec, Clément Calauzènes and Vianney Perchet.
• Bandits with side observations: bounded vs. logarithmic regret, UAI 2018,
  with Evrard Garcelon and Vianney Perchet.
• Combinatorial semi-bandit with known covariance, NIPS 2016,
  with Vianney Perchet.
• Anytime optimal algorithms in stochastic multi-armed bandits, ICML 2016
  with Vianney Perchet.

Teaching

Centrale Lille

• Fall 2023 - Sequential learning
• Fall 2022 - Sequential learning

Université de Lille

• Spring 2023 - Science des données 3, L3 MIASHS
• Spring 2022 - Science des données 3, L3 MIASHS

ENS Paris-Saclay (Master MVA)

• Spring 2023 - Sequential learning, with Pierre Gaillard
• Spring 2022 - Sequential learning, with Pierre Gaillard
• Spring 2021 - Sequential learning, with Pierre Gaillard

Université Paris Diderot

• Spring 2018 - Probabilités, L2 Math.
• Fall 2018 - Equation différentielles pour la biologie.
• Spring 2017 - Probabilités, L2 Math-Info.
• Fall 2017 - Raisonnement mathématique, L1 Informatique, L1 MIASHS.
• Spring 2016 - Analyse et algèbre 2, L1 Physique.