{"id":"arxiv/2607.03168","url":"https://arxiv.org/abs/2607.03168","tags":["math.OC","cs.LG"],"score":0,"title":"Entropy Regularization Improves Policy Robustness in Continuous-Time Reinforcement Learning","domain":"math.OC","source":"arxiv","authors":["Jialun Cao","Fernando Acero","David Šiška","Yufei Zhang"],"reviews":0,"subject":"math.OC","summary":"Entropy regularization is widely used in continuous-time reinforcement learning (RL) to reduce sensitivity to environmental perturbations, yet its robustness benefits lack a rigorous theoretical foundation. This paper establishes the first robustness guarantees for entropy-regularized continuous-time Markov decision processes. We show that maximizing an entropy-regularized objective yields a lower bound on a worst-case robust RL problem with joint reward and transition perturbations. We analytically characterize the induced robust sets and prove that they expand monotonically with the…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-07-03","reactions":{}}