List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on DMControl cheetah-run (expert)

96.28Normalized ScoreCQL + S2P
5.425629.012852.676.1872Sep 30, 2022Apr 12, 2023Oct 24, 2023May 5, 2024Nov 16, 2024May 29, 2025Dec 10, 2025
Updated 8mo ago

Evaluation results

MethodLinks
CQL + S2PS2P Augmentation=true2022.0996.28
CQLS2P Augmentation=false2022.0994.2
IQL + S2PS2P Augmentation=true2022.0987.18
IDM-TD3+BCaction-conditioned data percentage=5%2025.1285.6
TD3+BCaction-conditioned data percentage=100%, variant=Oracle2025.1283.4
IQLS2P Augmentation=false2022.0979.89
C-LAPaction-conditioned data percentage=100%, variant=Oracle2025.1279.6
C-LAPaction-conditioned data percentage=5%2025.1269.1
TD3+BCaction-conditioned data percentage=5%2025.1252.9
LAWMaction-conditioned data percentage=5%2025.1252.4
SLAC-off + S2PS2P Augmentation=true2022.0914.41
SLAC-offS2P Augmentation=false2022.098.92