Provably Safe Model-Based Meta Reinforcement Learning: An Abstraction-Based Approach

Re-assign community