Fetching the paper…

Learning Self-Correctable Policies and Value Functions from Demonstrations with Negative Sampling · Around