List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on MuJoCo HalfCheetah Friction shift

72.13Normalized ScoreCEDGE-Planner
6.422823.481440.5457.5986May 24, 2026
Updated 5mo ago

Evaluation results

MethodLinks
CEDGE-PlannerLevel=0.52026.0572.13
DiffuserLevel=0.52026.0571.25
IQLLevel=0.52026.0569.8
MOBODYLevel=0.52026.0569.54
BOSALevel=0.52026.0568.93
CEDGE-PolicyLevel=5.02026.0568.56
DARALevel=0.52026.0564.89
CEDGE-PolicyLevel=0.12026.0564.31
MOBODYLevel=5.02026.0559.2
MOBODYLevel=0.12026.0557.53
CEDGE-PolicyLevel=0.52026.0552.06
MOBODYLevel=2.02026.0550.02
TD3-BCLevel=0.52026.0549.43
CEDGE-PolicyLevel=2.02026.0549.15
CEDGE-PlannerLevel=2.02026.0547.93
DiffuserLevel=2.02026.0546.91
BOSALevel=2.02026.0546.53
DARALevel=2.02026.0546.25
IQLLevel=2.02026.0546.04
IQLLevel=5.02026.0544.96
BOSALevel=5.02026.0544.07
TD3-BCLevel=2.02026.0543.51
DARALevel=5.02026.0540.06
TD3-BCLevel=5.02026.0535.83
CEDGE-PlannerLevel=5.02026.0532.79
DiffuserLevel=5.02026.0532.11
IQLLevel=0.12026.0526.39
DARALevel=0.12026.0523.69
BOSALevel=0.12026.0512.53
DiffuserLevel=0.12026.0510.7
CEDGE-PlannerLevel=0.12026.0510.55
TD3-BCLevel=0.12026.058.95