ResearchBenchmarksLocomotion on MuJoCo Halfcheetah t = T/2 (shift)Follow1,526.7Average ReturnDQUCB1,405.4361,436.9181,468.41,499.882Oct 3, 2025Evaluation ResultsMethodMethodLinksAverage ReturnDQUCBT=200000, Action Bins...T=200000, Action Bins (K)=1002025.101,526.7DQN-UCBT=200000, Action Bins...T=200000, Action Bins (K)=1002025.101,526.1DQNT=200000, Action Bins...T=200000, Action Bins (K)=1002025.101,410.1