List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on MuJoCo Walker2d Gravity shift

69.48Normalized ScoreCEDGE-Policy
1.703219.299136.89554.4909May 24, 2026
Updated 5mo ago

Evaluation results

MethodLinks
CEDGE-PolicyLevel=0.12026.0569.48
MOBODYLevel=0.12026.0565.85
CEDGE-PolicyLevel=5.02026.0554.01
CEDGE-PlannerLevel=0.52026.0547.84
MOBODYLevel=5.02026.0546.05
CEDGE-PolicyLevel=0.52026.0544.36
CEDGE-PolicyLevel=2.02026.0544.36
MOBODYLevel=2.02026.0544.32
MOBODYLevel=0.52026.0543.57
IQLLevel=0.52026.0542.05
BOSALevel=0.52026.0540.09
TD3-BCLevel=0.12026.0536.48
CEDGE-PlannerLevel=0.12026.0534.08
DARALevel=2.02026.0532.2
DARALevel=0.52026.0529.72
TD3-BCLevel=0.52026.0527.43
DiffuserLevel=0.52026.0526.22
IQLLevel=2.02026.0525.69
DARALevel=0.12026.0520.12
DiffuserLevel=0.12026.0519.78
BOSALevel=0.12026.0518.75
IQLLevel=0.12026.0516.04
TD3-BCLevel=2.02026.0511.88
BOSALevel=2.02026.058.91
DiffuserLevel=2.02026.058.46
CEDGE-PlannerLevel=2.02026.056.97
DARALevel=5.02026.055.44
IQLLevel=5.02026.055.42
CEDGE-PlannerLevel=5.02026.055.36
BOSALevel=5.02026.055.25
TD3-BCLevel=5.02026.055.12
DiffuserLevel=5.02026.054.31