ResearchBenchmarksContinuous Control on Walker2d v1 (train)Follow4,682.82Max ReturnTD31,078.05522,013.90762,949.763,885.6124Feb 26, 2018Evaluation ResultsMethodMethodLinksMax ReturnTD32018.024,682.82PPO2018.023,317.69Our DDPG2018.023,098.11TRPO2018.022,321.47DDPG2018.021,843.85SAC2018.021,283.67ACKTR2018.021,216.7