Continuous Control on OpenAI Gym Mujoco 200K steps v2 (train)
1,000InvertedPendulum-v2 ReturnSSPG
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| SSPGUTD ratio=10, Random seeds=5, Test rollouts=1002022.10 | 1,000 | 3,487 | 4,793 | 10,309 | 5,513 | 5,148 | |
| REDQUTD ratio=20, Random seeds=5, Test rollouts=1002022.10 | 1,000 | 3,060 | 2,969 | 6,633 | 3,792 | 4,721 | |
| IAPORandom seeds=5, Test rollouts=1002022.10 | 1,000 | 426 | 570 | 5,303 | 118 | 390 | |
| SAC-20UTD ratio=20, Random seeds=5, Test rollouts=1002022.10 | 1,000 | 3,356 | 3,039 | 7,187 | 1,801 | 405 | |
| MBPOUTD ratio=20, Random seeds=5, Test rollouts=1002022.10 | 963 | 3,303 | 4,034 | 10,670 | 4,309 | 3,316 |