List your company
ResearchFormerly WizWand

Locomotion on HalfCheetah (Average episode returns)

1,483Average Episode ReturnPAWS (Trans.)
658.28872.391,086.51,300.61Jun 10, 2026
Updated 4mo ago

Evaluation results

MethodLinks
PAWS (Trans.)#Preferences=5002026.061,483
PAWS (MLP)#Preferences=5002026.061,450
IPL#Preferences=5002026.061,118
BC#Preferences=5002026.061,085
PAWS (Trans.)#Preferences=502026.061,063
P-IQL#Preferences=5002026.061,031
P-IQL#Preferences=502026.061,029
BC#Preferences=502026.061,019
CPL#Preferences=502026.06998
CPL+KL#Preferences=502026.06998
Pref Trans.#Preferences=5002026.06968
CPL#Preferences=5002026.06950
CPL+KL#Preferences=5002026.06948
Pref Trans.#Preferences=502026.06945
PAWS (MLP)#Preferences=502026.06810
IPL#Preferences=502026.06690