List your company
ResearchFormerly WizWand

Reinforcement Learning on DMControl reacher-easy (mixed)

75.61Avg Normalized ScoreIQL+S2P
61.965265.507669.0572.5924Sep 30, 2022
Updated 8mo ago

Evaluation results

MethodLinks
IQL+S2PBase Algorithm=IQL, S2P Augmentation=true2022.0975.61
CQL+S2PBase Algorithm=CQL, S2P Augmentation=true2022.0975.53
CQLBase Algorithm=CQL, S2P Augmentation=false2022.0970.37
IQLBase Algorithm=IQL, S2P Augmentation=false2022.0966.88
SLAC-off+S2PBase Algorithm=SLAC-off, S2P Augmentation=true2022.0963.59
SLAC-offBase Algorithm=SLAC-off, S2P Augmentation=false2022.0962.49