ResearchBenchmarksGoal-conditioned Reinforcement Learning on DeepMind Control Suite P2E tasks standard (test)Follow968Walker Stand ScoreDrQv2305.52477.51649.5821.49Oct 18, 2021Evaluation ResultsMethodMethodLinksWalker Stand ScoreHopper Stand ScoreCartpole Balance ScoreCartpole Balance (Sparse)Pendulum Swing Up ScoreCup Catch ScoreReacher Hard ScoreDrQv2Zero-Shot=false, Oracl...Zero-Shot=false, Oracle=true2021.10968957989983837909970LEXAZero-Shot=true, Early...Zero-Shot=true, Early stopping=true2021.10957840886996788969937P2EZero-Shot=trueZero-Shot=true2021.1033184195086079296266