ResearchBenchmarksLocomotion on MuJoCo Halfcheetah (t = 2T/3 (shift))Follow2,017.6Average ReturnDQUCB1,578.721,692.661,806.61,920.54Oct 3, 2025Evaluation ResultsMethodMethodLinksAverage ReturnDQUCBT=200000, Action Bins...T=200000, Action Bins (K)=1002025.102,017.6DQN-UCBT=200000, Action Bins...T=200000, Action Bins (K)=1002025.101,733.4DQNT=200000, Action Bins...T=200000, Action Bins (K)=1002025.101,595.6