List your company
ResearchFormerly WizWand

Reinforcement Learning on DMControl ball-in-cup-catch (random)

92.81Normalized ScoreCQL+S2P
69.129275.277181.42587.5729Sep 30, 2022
Updated 8mo ago

Evaluation results

MethodLinks
CQL+S2PBase Algorithm=CQL, S2P Augmentation=true2022.0992.81
SLAC-off+S2PBase Algorithm=SLAC-off, S2P Augmentation=true2022.0985.57
CQLBase Algorithm=CQL, S2P Augmentation=false2022.0984.87
IQL+S2PBase Algorithm=IQL, S2P Augmentation=true2022.0982.28
IQLBase Algorithm=IQL, S2P Augmentation=false2022.0974.77
SLAC-offBase Algorithm=SLAC-off, S2P Augmentation=false2022.0970.04