Fetching the paper…

RePo: Resilient Model-Based Reinforcement Learning by Regularizing Posterior Predictability · Around