Rémi Leluc

Rémi Leluc

Quantitative Researcher

Rémi Leluc, PhD

About me

I am a Quantitative Researcher at Qube Research and Technologies (QRT) working on quantitative strategies for portfolio optimization. Prior to that, I was a postdoctoral researcher in the Department of Applied Mathematics (CMAP) at École Polytechnique working with Aymeric Dieuleveut on Federated Learning and (stochastic) optimization problems. I obtained my PhD in Applied Mathematics and Computer Science from Institut Polytechnique de Paris at Télécom Paris in the team S2A (Signal, Statistics and Learning) of LTCI.

I conducted my PhD under the supervision of François Portier and Pascal Bianchi. You can find the manuscript on theses.fr and the slides of the defense here. Before that, I graduated from Télécom Paris in 2019. I hold a master’s degree in applied mathematics from Télécom Paris and a master’s degree in Mathematics, Vision and Learning (MVA) from Ecole Normale Supérieure Paris-Saclay.

For more details, please see my CV.

My Research

My research topics/interests focus on:

  • Monte Carlo methods
  • Stochastic Optimization
  • Reinforcement Learning
  • Federated Learning

More generally, I am interested in applications of machine learning to real-world problems such as building intelligent systems that enable collaborative learning. I am always looking for collaborations and would love to hear from you! Reach out if any of my work sounds interesting.

News

Archive

  • 05.2024 : “Sliced-Wasserstein Estimation with Spherical Harmonics as Control Variates” is accepted to ICML2024. Congrats to A. Dieuleveut, F. Portier, J. Segers and A. Zhuman!
  • 04.2024 : “Speeding up Monte Carlo Integration: Control Neighbors for Optimal Convergence” is accepted to Bernoulli. With F. Portier, J. Segers and A. Zhuman.
  • 04.2024: I attended the “25th Anniversary of Eurandom” (link) in Eindhoven (Netherlands), and gave a talk there on “Monte Carlo Methods in Machine Learning”.
  • 01.2024 : “Compression with Exact Error Distribution for Federated Learning” is accepted to AISTATS2024. With M. Hegazy, C.T. Li and A. Dieuleveut.
  • 08.2023 : “Asymptotic Analysis of Conditioned Stochastic Gradient Descent” is accepted to Transactions on Machine Learning Research 2023. With François Portier.
  • 04.2023 : I started a postdoc at Ecole Polytechnique with Aymeric Dieuleveut.
  • 03.2023 : I defended my PhD ! Slides
  • 10.2022 : “SGD with Coordinate Sampling: Theory and Practice” is accepted to Journal of Machine Learning Research 2022. With François Portier.
  • 09.2022 : “A Quadrature Rule combining Control Variates and Adaptive Importance Sampling” is accepted to Neurips2022. With François Portier, Johan Segers and Aigerim Zhuman.
  • 24.07/30.07 2022: I attended the “Math for Machine Learning Summer School” (link) at the Mohammed VI university, Ben Guerir, Morocco, and gave a talk there.
  • 10.2021 - 04.2022: I worked as an Artificial Intelligence Researcher at TotalEnergies OneTech in the Data AI team of Sébastien Gourvénec to study RL techniques in an industrial environment. Filing of a patent
  • 05.2021: “Feature Clustering for Support Identification in Extreme Regions” accepted to ICML2021. With Hamid Jalalzai.
  • 04.2021: “Control Variate Selection for Monte Carlo Integration” accepted to Statistics and Computing. With François Portier and Johan Segers.
  • 28.08/10.07 2020: I attended Machine Learning Summer School (MLSS 2020, competitive selection process with 15% rate of application acceptance)