Excited to share our latest work “Provable and Practical: Efficient Exploration in Reinforcement Learning via Langevin Monte Carlo” w/ @LanceLan3, @iampanxu, A. Rupam Mahmood, Doina Precup, @AnimaAnandkumar and @Azizzadenesheli.
Paper: arxiv.org/abs/2305.18246🧵👇