List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on Hopper Medium-Replay JointNoise Shift

93.704Average Return1T10S
39.7446453.7533267.76281.77068Oct 30, 2024
Updated 8mo ago

Evaluation results

MethodLinks
1T10SFramework=QT2024.1093.704
QTAugmentation Strategy=1T10S2024.1093.704
REAG*MVFramework=QT2024.1093.409
QTAugmentation Strategy=REAG_MV2024.1093.409
REAG*DaraFramework=DT2024.1083.525
DTAugmentation Strategy=REAG_Dara2024.1083.525
REAG*MVFramework=DT2024.1077.825
DTAugmentation Strategy=REAG_MV2024.1077.825
1T10SFramework=DT2024.1061.87
DTAugmentation Strategy=1T10S2024.1061.87
REAG*DaraFramework=Reinformer2024.1052.052
ReinformerAugmentation Strategy=REAG_Dara2024.1052.052
REAG*DaraFramework=QT2024.1051.456
QTAugmentation Strategy=REAG_Dara2024.1051.456
REAG*MVFramework=Reinformer2024.1043.985
ReinformerAugmentation Strategy=REAG_MV2024.1043.985
1T10SFramework=Reinformer2024.1041.82
ReinformerAugmentation Strategy=1T10S2024.1041.82