ResearchBenchmarksContinuous Control on Hopper v1 (train)Follow3,564.07Max ReturnTD31,791.8582,251.95152,712.0453,172.1385Feb 26, 2018Evaluation ResultsMethodMethodLinksMax ReturnTD32018.023,564.07SAC2018.022,996.66TRPO2018.022,471.3ACKTR2018.022,428.39PPO2018.022,164.7DDPG2018.022,020.46Our DDPG2018.021,860.02