List your company

Offline Reinforcement Learning on D4RL Gym walker2d medium

102.4Normalized ReturnRORL
-4.40823.32151.0578.779Oct 1, 2021Jun 27, 2022Mar 23, 2023Dec 17, 2023Sep 11, 2024Jun 7, 2025Mar 4, 2026

Evaluation results

MethodLinks
RORLImplementation Source=Ours, Ensemble Size=102022.06102.4
SPQRBase algorithm=SAC-Min2024.0198.4
SPQRBase algorithm=EDAC2024.0194.8
EDACImplementation=Ours2021.1092.5
EDACImplementation Source=Paper2022.0692.5
EDAC2024.0192.5
PBRL2022.0689.6
IPDEvaluation Protocol=Tenfold Episode Evaluation2026.0389.5
GTP2025.1089.5
SAC-NImplementation=Ours2021.1087.9
SAC-Min2024.0187.9
EDAC-10Implementation Source=Reproduced, Ensemble Size=102022.0687.6
QTEvaluation Protocol=Tenfold Episode Evaluation2026.0387.6
QIPO-OT2025.1087.6
Repb-SDE2021.1085.3
Repb-SDEalgorithm class=model-based2021.1085.3
QQLDomain=Gym Locomotion, Hyperparameter Tuning=consistent2025.1185.2
QIPO-Diff2025.1085
ROMI-CQL2021.1084.3
ROMI-CQLbase algorithm=CQL2021.1084.3
XQLDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.1184.2
SSCQL2025.1084.2
CQL2021.1083.8
CQLalgorithm class=model-free2021.1083.8
MXQLDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.1183.8
TD3+BCDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.1183.7
CQLEvaluation Protocol=Tenfold Episode Evaluation2026.0383
DDEvaluation Protocol=Tenfold Episode Evaluation2026.0382.5
ReinformerEvaluation Protocol=Tenfold Episode Evaluation2026.0380.5
BRACImplementation=Reproduced2021.1079.6
CQLImplementation=Reproduced2021.1079.5
CQL2022.0679.5
IQLEvaluation Protocol=Tenfold Episode Evaluation2026.0378.3
IQLDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.1178.3
MOREL2021.1077.8
COMBO2021.1075.5
BCDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.1175.3
CQLImplementation=Original Paper2021.1074.5
CQL-Min2024.0174.5
DTEvaluation Protocol=Tenfold Episode Evaluation2026.0374
EDTEvaluation Protocol=Tenfold Episode Evaluation2026.0372.8
CQLDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.1172.5
ROMI-BCQ2021.1072.4
BCQImplementation=Reproduced2021.1071.8
BCImplementation=Reproduced2021.1070.9
BC2022.0670.9
BC2024.0170.9
BAIL2021.1068.8
QDTEvaluation Protocol=Tenfold Episode Evaluation2026.0367.1
FQL2025.1065.9
BEARImplementation=Reproduced2021.1059.8
MORELImplementation=Reproduced2021.1057
SAC-10Implementation Source=Reproduced, Ensemble Size=102022.0646.7
BCQ2021.1045.2
BC2021.1041.7
BC2021.1041.7
UWACImplementation=Reproduced2021.1016
REMImplementation=Reproduced2021.100.2
BRAC-P2021.100.1
MOPO2021.10-0.2
SACImplementation=Reproduced2021.10-0.3
BEAR2021.10-0.3
BRAC-v2021.10-0.3