List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on D4RL Gym halfcheetah-random

35.4Normalized ScoreCQL-Min
0.8729.83618.827.764Jun 6, 2022Jan 30, 2023Sep 25, 2023May 21, 2024Jan 14, 2025Sep 9, 2025May 6, 2026
Updated 5mo ago

Evaluation results

MethodLinks
CQL-Min2024.0135.4
SPQRBase algorithm=SAC-Min2024.0133.5
DOSER2026.0532.8
CQL2022.0631.3
SAC-10Implementation Source=Reproduced, Ensemble Size=102022.0629
DMG2026.0528.8
RORLImplementation Source=Ours, Ensemble Size=102022.0628.5
EDACImplementation Source=Paper2022.0628.4
EDAC2024.0128.4
SAC-Min2024.0128
CQL2026.0517.5
EDAC-10Implementation Source=Reproduced, Ensemble Size=102022.0613.4
IQL2026.0513.1
PBRL2022.0611
TD3+BC2026.0511
AWAC2026.056.1
BC2026.052.6
BEAR2026.052.3
OneStep2026.052.3
BC2022.062.2
BC2024.012.2
BCQ2026.052.2
DT2026.052.2