Verifiable Reinforcement Learning via Policy Extraction
2018·Arxiv