ResearchBenchmarksOffline Goal-Conditioned Reinforcement Learning on D'ClawTurn (offline)For AI assistantsMarkdownContent indexOffline Goal-Conditioned Reinforcement Learning on D'ClawTurn (offline)Follow9.34Discounted ReturnGoFAR-0.37362.14824.677.1918Jun 7, 2022Evaluation resultsMethodMethodLinksDiscounted ReturnGoFAR2022.069.34AMLearning Paradigm=Acto...Learning Paradigm=Actor-Critic, HER rate=1.02022.062.82WGCSLLearning Paradigm=Supe...Learning Paradigm=Supervised Learning, HER rate=1.02022.060GCSLLearning Paradigm=Supe...Learning Paradigm=Supervised Learning, HER rate=1.02022.060DDPGLearning Paradigm=Acto...Learning Paradigm=Actor-Critic, HER rate=0.22022.060