ResearchBenchmarksLocomotion on MuJoCo Hopper t = T/2 (shift)Follow211.8Average ReturnDQUCB178.104186.852195.6204.348Oct 3, 2025Evaluation ResultsMethodMethodLinksAverage ReturnDQUCBT=200000, Action Bins...T=200000, Action Bins (K)=1002025.10211.8DQN-UCBT=200000, Action Bins...T=200000, Action Bins (K)=1002025.10211DQNT=200000, Action Bins...T=200000, Action Bins (K)=1002025.10179.4