List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on D4RL Gym halfcheetah-medium-expert

114Normalized ReturnSPQR
-3.83226.75957.3587.941Oct 1, 2021Jun 9, 2022Feb 15, 2023Oct 24, 2023Jul 1, 2024Mar 9, 2025Nov 15, 2025
Updated 5mo ago

Evaluation results

MethodLinks
SPQRBase algorithm=SAC-Min2024.01114
RORLImplementation Source=Ours, Ensemble Size=102022.06107.8
EDAC-10Implementation Source=Reproduced, Ensemble Size=102022.06107.2
SAC-NImplementation=Ours2021.10107.1
SAC-10Implementation Source=Reproduced, Ensemble Size=102022.06107.1
SAC-Min2024.01107.1
EDACImplementation=Ours2021.10106.3
EDACImplementation Source=Paper2022.06106.3
EDAC2024.01106.3
SSCQL2025.1098.1
MORELalgorithm class=model-based2021.1095.6
BCQImplementation=Reproduced2021.1095.4
CQLImplementation=Reproduced2021.1095
CQL2022.0695
QIPO-OT2025.1094.4
QQLDomain=Gym Locomotion, Hyperparameter Tuning=consistent2025.1194.3
XQLDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.1194.2
QIPO-Diff2025.1094.1
GTP2025.1093.8
PBRL2022.0692.3
MXQLDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.1192.1
FQL2025.1092
CQLDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.1191.6
CQLalgorithm class=model-free2021.1090.8
TD3+BCDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.1190.7
ROMI-CQLbase algorithm=CQL2021.1086.8
IQLDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.1186.7
MORELImplementation=Reproduced2021.1080.4
CQLImplementation=Original Paper2021.1062.4
CQL-Min2024.0162.4
BCDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.1155.2
BRACImplementation=Reproduced2021.1052.3
UWACImplementation=Reproduced2021.1047
BEARImplementation=Reproduced2021.1045.7
BCImplementation=Reproduced2021.1044
BC2022.0644
BC2024.0144
BC2021.1039.7
SACImplementation=Reproduced2021.1028.4
REMImplementation=Reproduced2021.100.7