ResearchBenchmarksContinuous Control on Ant v1 (train)Follow4,372.44Max Average ReturnTD3-253.7816947.25672,148.2953,349.3333Feb 26, 2018Evaluation ResultsMethodMethodLinksMax Average ReturnTD32018.024,372.44ACKTR2018.021,821.94PPO2018.021,083.2DDPG2018.021,005.3Our DDPG2018.02888.77SAC2018.02655.35TRPO2018.02-75.85