ResearchBenchmarksLocomotion on MuJoCo Hopper t = 2T/3 (shift)Follow397.5Average ReturnDQUCB213.94261.595309.25356.905Oct 3, 2025Evaluation ResultsMethodMethodLinksAverage ReturnDQUCBT=200000, Action Bins...T=200000, Action Bins (K)=1002025.10397.5DQN-UCBT=200000, Action Bins...T=200000, Action Bins (K)=1002025.10257.2DQNT=200000, Action Bins...T=200000, Action Bins (K)=1002025.10221