Lucas Lévy

I am a research intern in the SIERRA project-team at INRIA Paris, working under the supervision of Michael I. Jordan and Francis Bach. My research interests lie at the intersection of Optimization and Machine Learning Theory.

I recently earned my Engineering Degree from Ecole polytechnique (X 2022), specializing in Applied Mathematics, alongside the MVA (Mathematics, Vision, Learning) Master's degree at ENS Paris-Saclay. Have a look at my CV for additional information.

I was also a visiting research student at the University of Oxford, where I worked on online optimization under bandit feedback under the supervision of Arya Akhavan and Patrick Rebeschini.

Publications

Self-Concordant Perturbations for Linear Bandits
Lucas Lévy, Jean-Lou Valeau, Arya Akhavan, Patrick Rebeschini
Conference on Learning Theory (COLT), 2026
[arxiv]

News

🇦🇹 I will be giving a 25-minute invited talk at the IMS Annual Meeting 2026 in Salzburg. Title: Sampling-Based Regularisation for Adversarial Linear Bandits.
July 6, 2026

🌎 I will be attending the 39th Conference on Learning Theory (COLT) in San Diego to present our work on self-concordant perturbations. [slides] [poster]
June 29 - July 3, 2026

📄 I am excited to share that my first first-author paper has been accepted to COLT, based on research conducted as a research student at University of Oxford. In this paper, we analyze how exploration in bandit problems can be achieved through self-concordance and perturbation-based regularization.
May 2026

📍 I started my internship at INRIA Paris, where I am fortunate to be advised by Michael I. Jordan and Francis Bach.
April 2026