List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on 1T10S Hopper (Medium-Expert)

111.587ScoreD-BCQ
19.3878843.3241967.260591.19681Oct 30, 2024
Updated 8mo ago

Evaluation results

MethodLinks
D-BCQDistribution Shift=JointNoise2024.10111.587
D-BCQDistribution Shift=BodyMass2024.10110.014
H2ODistribution Shift=JointNoise2024.10109.457
BCQDistribution Shift=BodyMass2024.10109.367
H2ODistribution Shift=BodyMass2024.10108.872
BCQDistribution Shift=JointNoise2024.10108.639
IGDFDistribution Shift=BodyMass2024.10105.465
IGDFDistribution Shift=JointNoise2024.1099.057
D-CQLDistribution Shift=JointNoise2024.1094.128
D-AWRDistribution Shift=BodyMass2024.1078.262
AWRDistribution Shift=JointNoise2024.1074.708
DFDTDistribution Shift=JointNoise2024.1074.306
CQLDistribution Shift=JointNoise2024.1072.512
D-CQLDistribution Shift=BodyMass2024.1072.149
DFDTDistribution Shift=BodyMass2024.1070.703
CQLDistribution Shift=BodyMass2024.1070.467
D-AWRDistribution Shift=JointNoise2024.1060.952
AWRDistribution Shift=BodyMass2024.1057.595
BEARDistribution Shift=JointNoise2024.1039.031
D-BEARDistribution Shift=JointNoise2024.1033.052
D-MOPODistribution Shift=JointNoise2024.1032.589
D-BEARDistribution Shift=BodyMass2024.1031.09
MOPODistribution Shift=BodyMass2024.1030.541
D-MOPODistribution Shift=BodyMass2024.1030.54
MOPODistribution Shift=JointNoise2024.1030.537
BEARDistribution Shift=BodyMass2024.1022.934