List your company
ResearchFormerly WizWand

Worst-case time-constrained reinforcement learning on Walker MuJoCo (test)

1.69Normalized Worst-Case RewardStacked-TC-M2TD3
-0.26520.24240.751.2576Jun 12, 2024
Updated 8mo ago

Evaluation results

MethodLinks
Stacked-TC-M2TD3Architecture=Stacked, Framework=TC2024.061.69
Oracle-TC-M2TD3Oracle=true, Framework=TC2024.061.49
TC-M2TD3Framework=TC2024.061.38
Oracle-TC-RARLOracle=true, Framework=TC2024.061.34
Oracle M2TD3Oracle=true2024.061.31
DRFramework=Domain Randomization2024.061.03
Stacked-TC-RARLArchitecture=Stacked, Framework=TC2024.061.01
M2TD3Framework=Robust2024.061
TC-RARLFramework=TC2024.060.89
RARLFramework=Robust2024.060.27
TD3Framework=Vanilla2024.060
Oracle RARLOracle=true2024.06-0.19