List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on Hopper kinematic shifts

97ScoreOTDF
44.27257.96171.6585.339Dec 2, 2025
Updated 8mo ago

Evaluation results

MethodLinks
OTDFSource Policy=expert, Target Policy=expert2025.1297
DROCOSource Policy=expert, Target Policy=expert2025.1289.3
DROCOSource Policy=medium, Target Policy=expert2025.1280.8
IGDFSource Policy=expert, Target Policy=expert2025.1270.1
OTDFSource Policy=medium, Target Policy=expert2025.1269.3
IQLSource Policy=expert, Target Policy=expert2025.1262.6
DROCOSource Policy=expert, Target Policy=medium2025.1262.2
IGDFSource Policy=medium, Target Policy=expert2025.1261.8
IGDFSource Policy=expert, Target Policy=medium2025.1261.3
IQLSource Policy=medium, Target Policy=expert2025.1256.1
DROCOSource Policy=medium, Target Policy=medium2025.1255.4
IGDFSource Policy=medium, Target Policy=medium2025.1254.3
IQLSource Policy=expert, Target Policy=medium2025.1253.6
OTDFSource Policy=expert, Target Policy=medium2025.1251.4
IQLSource Policy=medium, Target Policy=medium2025.1248.8
OTDFSource Policy=medium, Target Policy=medium2025.1246.3