List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on D4RL MuJoCo Walker2d medium-expert v2

117.2Average Normalized ScoreCBOP
-4.06427.41858.990.382Jun 7, 2023Nov 21, 2023May 7, 2024Oct 22, 2024Apr 7, 2025Sep 22, 2025Mar 9, 2026
Updated 7mo ago

Evaluation results

MethodLinks
CBOP2024.05117.2
MOBILE2024.05115.2
ADMPO-OFF2024.05114.9
EDAC2024.05114.7
ROMItraining steps=1M, seeds=5, version=v22026.03113.3
MOPO2024.05112.9
IQLtraining steps=1M, seeds=5, version=v22026.03110.7
DOGERatio=1, Size=2M2023.06110.4
COUNTtraining steps=1M, seeds=5, version=v22026.03110.4
TD3+BCRatio=1, Size=2M2023.06110.1
TD3+BC2024.05110.1
TSRLRatio=1, Size=2M2023.06109.8
IQLRatio=1, Size=2M2023.06109.6
CQL2024.05109.1
CQLRatio=1, Size=2M2023.06108.8
CQLtraining steps=1M, seeds=5, version=v22026.03107.9
BCRatio=1, Size=2M2023.06107.5
COMBO2024.05103.3
MOBILEtraining steps=1M, seeds=5, version=v22026.03103.2
MOPOtraining steps=1M, seeds=5, version=v22026.0393.8
BC2024.0590.1
RAMBOtraining steps=1M, seeds=5, version=v22026.0373.7
RAMBO2024.0556.7
TSRLRatio=1/200, Size=10k2023.0646.4
MOPORatio=1, Size=2M2023.0644.6
DOGERatio=1/200, Size=10k2023.0635.3
IQLRatio=1/200, Size=10k2023.0626.5
BCRatio=1/200, Size=10k2023.0621.7
CQLRatio=1/200, Size=10k2023.0619.1
TD3+BCRatio=1/200, Size=10k2023.067.9
MOPORatio=1/200, Size=10k2023.060.6