List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on D4RL MuJoCo Hopper-medium-v2

107.4Avg Normalized ScoreADMPO-OFF
1.42428.93756.4583.963Jun 7, 2023Nov 21, 2023May 7, 2024Oct 22, 2024Apr 7, 2025Sep 22, 2025Mar 9, 2026
Updated 7mo ago

Evaluation results

MethodLinks
ADMPO-OFF2024.05107.4
MOBILE2024.05106.6
ROMItraining steps=1M, seeds=5, version=v22026.03105
CBOP2024.05102.6
EDAC2024.05101.6
COUNTtraining steps=1M, seeds=5, version=v22026.03100.2
DOGERatio=1, Size=1M2023.0698.6
COMBO2024.0597.2
RAMBOtraining steps=1M, seeds=5, version=v22026.0395.2
RAMBO2024.0587
TSRLRatio=1, Size=1M2023.0686.7
MOBILEtraining steps=1M, seeds=5, version=v22026.0377.6
MOPOtraining steps=1M, seeds=5, version=v22026.0372.2
IQLRatio=1, Size=1M2023.0666.3
IQLtraining steps=1M, seeds=5, version=v22026.0366.2
MOPO2024.0562.8
TSRLRatio=1/100, Size=10k2023.0662
CQL2024.0561.9
TD3+BCRatio=1, Size=1M2023.0659.3
TD3+BC2024.0559.3
CQLRatio=1, Size=1M2023.0658.5
BC2024.0554.1
CQLtraining steps=1M, seeds=5, version=v22026.0353
BCRatio=1, Size=1M2023.0652.9
IQLRatio=1/100, Size=10k2023.0646.7
DOGERatio=1/100, Size=10k2023.0644.2
CQLRatio=1/100, Size=10k2023.0643.1
TD3+BCRatio=1/100, Size=10k2023.0640.1
BCRatio=1/100, Size=10k2023.0629.7
MOPORatio=1, Size=1M2023.0628
MOPORatio=1/100, Size=10k2023.065.5