List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on D4RL Gym walker2d (expert)

121.6Normalized Avg ReturnCQL-Min
-4.13628.50761.1593.793Oct 4, 2021Feb 18, 2022Jul 5, 2022Nov 20, 2022Apr 6, 2023Aug 21, 2023Jan 6, 2024
Updated 8mo ago

Evaluation results

MethodLinks
CQL-Min2024.01121.6
SAC-Min2024.01116.7
RORLImplementation Source=Ours, Ensemble Size=102022.06115.4
SPQRBase algorithm=SAC-Min2024.01115.2
EDACImplementation=Ours2021.10115.1
EDACImplementation Source=Paper2022.06115.1
EDAC2024.01115.1
CQLImplementation=Reproduced2021.10109.3
CQL2022.06109.3
BCImplementation=Reproduced2021.10108.7
BC2022.06108.7
BC2024.01108.7
PBRL2022.06108.3
BEARImplementation=Reproduced2021.10107.7
SAC-NImplementation=Ours2021.10107.4
CQLImplementation=Original Paper2021.10107
BCQImplementation=Reproduced2021.10106.3
EDAC-10Implementation Source=Reproduced, Ensemble Size=102022.0657.8
MORELImplementation=Reproduced2021.1040.1
BRACImplementation=Reproduced2021.1012.2
UWACImplementation=Reproduced2021.101.5
SAC-10Implementation Source=Reproduced, Ensemble Size=102022.061.2
SACImplementation=Reproduced2021.100.7
REMImplementation=Reproduced2021.100.7