Fetching the paper…

Learning from Suboptimal Demonstration via Self-Supervised Reward Regression · Around