Policy Optimization on MuJoCo HalfCheetah H=40
49.1ReturnMAX-RETURN
Evaluation Results
| Method | Links | |
|---|---|---|
| MAX-RETURNHorizon=40, Optimization Algorithm=SAC2022.06 | 49.1 | |
| IPWHorizon=40, Optimization Algorithm=SAC2022.06 | -11.3 | |
| SCIGANHorizon=40, Optimization Algorithm=SAC2022.06 | -11.4 | |
| OFF-SLHorizon=40, Optimization Algorithm=SAC2022.06 | -13.2 | |
| GALILEOHorizon=40, Optimization Algorithm=SAC2022.06 | -14.2 |