List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on Walker2d D4RL v2 (offline)

86.7ReturnTRACER
-3.88419.63343.1566.667Nov 1, 2024
Updated 8mo ago

Evaluation results

MethodLinks
TRACERCorrupted Element=action2024.1186.7
TRACERCorrupted Element=reward2024.1185.5
RIQLCorrupted Element=action2024.1184.6
EDACCorrupted Element=action2024.1183.2
RIQLCorrupted Element=reward2024.1183.2
RIQLCorrupted Element=dynamics2024.1178.2
TRACERCorrupted Element=dynamics2024.1175.9
IQLCorrupted Element=action2024.1171.3
CQLCorrupted Element=reward2024.1169.4
IQLCorrupted Element=reward2024.1165.3
CQLCorrupted Element=action2024.1162.7
UWMSGCorrupted Element=reward2024.1162
TRACERCorrupted Element=observation2024.1132.7
UWMSGCorrupted Element=action2024.1131.5
RIQLCorrupted Element=observation2024.1128.4
IQLCorrupted Element=observation2024.1127.2
MSGCorrupted Element=action2024.1125.3
CQLCorrupted Element=observation2024.1119.4
MSGCorrupted Element=reward2024.1118.4
BCCorrupted Element=action2024.1118.1
IQLCorrupted Element=dynamics2024.1117.7
BCCorrupted Element=reward2024.1116
BCCorrupted Element=dynamics2024.1116
BCCorrupted Element=observation2024.119.6
MSGCorrupted Element=dynamics2024.117.4
UWMSGCorrupted Element=observation2024.116.2
EDACCorrupted Element=reward2024.114.3
UWMSGCorrupted Element=dynamics2024.110.2
EDACCorrupted Element=dynamics2024.11-0.1
EDACCorrupted Element=observation2024.11-0.2
CQLCorrupted Element=dynamics2024.11-0.2
MSGCorrupted Element=observation2024.11-0.4