ResearchBenchmarksRobotic Control on Ant MDPFollow5,614.1Max Average ReturnSAC271.50561,658.52533,045.5454,432.5647Sep 12, 2022Evaluation ResultsMethodMethodLinksMax Average Returnt-test P-value (PPO vs Baseline)t-test P-value (Original vs Baseline)SAC2022.095,614.14—MSACHistory Length=5History Length=52022.095,000.6244TD32022.094,773.74—LSTM-TD3History Length=5History Length=52022.094,070.0344MTD3History Length=5History Length=52022.093,236.9344PPO2022.09476.99——