Fetching the paper…

Safe Model-Based Reinforcement Learning with an Uncertainty-Aware Reachability Certificate · Around