SAC Temperature Tuning on Pendulum v1
-936Return @ 5k StepsFixed-T
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Fixed-TNumber of seeds=5, Return averaging method=mean ± std2026.06 | -936 | -408 | -440 | -237 | -279 | 0.5 | 0.716 | 0 | |
| Orig-SACNumber of seeds=5, Return averaging method=mean ± std2026.06 | -951 | -301 | -296 | -256 | -278 | 0.17 | 0.223 | 68 | |
| RF-BONumber of seeds=5, Return averaging method=mean ± std2026.06 | -996 | -320 | -276 | -200 | -251 | 0.17 | 0.174 | 52 |