High-Dimensional Control on SafetyPointGoal1 v0 (test)
28.25RewardSAC
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SACDiscounting Variant=AdaGamma2026.05 | 28.25 | 29.37 | |
| SACDiscounting Variant=Fixed-γ2026.05 | 27.82 | 52.53 | |
| SACDiscounting Variant=Uncertainty2026.05 | 27.47 | 37.46 | |
| SACDiscounting Variant=CrossValidate2026.05 | 27.34 | 45.84 | |
| PPODiscounting Variant=AdaGamma2026.05 | 26.31 | 41.82 | |
| PPODiscounting Variant=Uncertainty2026.05 | 22.59 | 51.25 | |
| PPODiscounting Variant=Fixed-γ2026.05 | 22.58 | 51.26 | |
| PPODiscounting Variant=CrossValidate2026.05 | 20.62 | 54.13 |