About Me
I am a Computer Science student from ENS Rennes specializing in Reinforcement Learning (RL) and sequential decision-making under uncertainty. I'm currently completing my final Master's internship at LIP6 (Sorbonne Université) on hardware-accelerated Multi-Objective and Multi-Agent Reinforcement Learning (MOMARL) benchmarking with JAX. Starting this fall, I will be pursuing a PhD with Pr. Aurélie Beynier on improving Multi-Objective and Multi-Agent RL
Current Research Focus
- NNs & POMDP: Developing theory-grounded methods to solve POMDPs with Neural Networks to achieve stronger theoretical guarantees at scale. Paper accepted at RDPIA 2026 (in french only for now). I presented this work during PFIA 2026 at Arras! Here you can find my presentation done with Manim-Slides. I will record a french audio for that when I have time :-)
- MB-MARL: Designing a Multi-Agent RL solution using an actor-critic approach to learn a decentralized world-model using a centralized one. While having the ability to handle a varying number of agents.