Fetching the paper…

SAFER: Data-Efficient and Safe Reinforcement Learning via Skill Acquisition · Around