Yonatan Ashlag — accepted AI-conference papers

2026

Probing in the Dark: State Entropy Maximization for POMDPs

Yonatan Ashlag, Mirco Mutti, Aviv Tamar, Kfir Yehuda Levy

ICLR 2026poster

Sample efficiency is one of the main bottlenecks for optimal decision making via reinforcement learning. Pretraining a policy to maximize the entropy of the state visitation can substantially speedup reinforcement learning of downstream tasks. It is still an open question how to maximize the state e…

Cited by 0 Source ↗code ↗Scholar ↗

2025

State Entropy Regularization for Robust Reinforcement Learning

Yonatan Ashlag, Uri Koren, Mirco Mutti, Esther Derman, Pierre-Luc Bacon, Shie Mannor

NeurIPS 2025oral

State entropy regularization has empirically shown better exploration and sample complexity in reinforcement learning (RL). However, its theoretical guarantees have not been studied. In this paper, we show that state entropy regularization improves robustness to structured and spatially correlated p…

Cited by 0 Source ↗Scholar ↗