My research lies in applied probability, with a focus on reinforcement learning (RL) algorithms in hard-exploration environments — settings where reward signals are sparse or highly stochastic. I study these problems through the framework of rare event estimation and simulation, using tools from stochastic processes, measure changes via exponential martingales, large deviation principles, and Feynman–Kac semigroups.

PhD thesis (in progress)

Exploration in reinforcement learning with scarce resources
Cotutelle: LAAS-CNRS / Université de Toulouse III & PEDECIBA / Universidad de la República.
Supervisors: Paola Bermolen (Uruguay), Matthieu Jonckheere (France).
Expected defense: October 2026.

Research supervision

Research intern co-supervision · LAAS-CNRS / Toulouse INP, April–June 2024
Co-supervised with Dr. Matthieu Jonckheere. Over seven weeks, guided Malik Hacini in a research project on generalized spectral clustering. The internship resulted in an open-source repository (github.com/Malik-Hacini/Generalized-Spectral-Clustering ) and a written monograph presented at Toulouse INP.

IMERL, Faculty of Engineering, Universidad de la República

Teaching assistant (Ayudante G1 / Asistente G2) · July 2019 – December 2022