Continuous Control on MuJoCo HalfCheetah v5
12,332.38Max ReturnFMER
Evaluation Results
| Method | Links | |
|---|---|---|
| FMEREnvironment steps=1M, Random seeds=52026.03 | 12,332.38 | |
| DPMDEnvironment steps=1M, Random seeds=52026.03 | 10,706.06 | |
| QVPOEnvironment steps=1M, Random seeds=52026.03 | 10,309.6 | |
| SACEnvironment steps=1M, Random seeds=52026.03 | 10,295.33 | |
| DIPOEnvironment steps=1M, Random seeds=52026.03 | 9,516.47 | |
| TD3Environment steps=1M, Random seeds=52026.03 | 9,496.59 | |
| PPOEnvironment steps=1M, Random seeds=52026.03 | 5,982.74 | |
| FPOEnvironment steps=1M, Random seeds=52026.03 | 2,860.01 | |
| PPOTraining steps=1,000,000, Seeds=102026.05 | 2,363 | |
| TQCTraining steps=1,000,000, Seeds=102026.05 | 12 | |
| TD3Training steps=1,000,000, Seeds=102026.05 | 10 | |
| SACTraining steps=1,000,000, Seeds=102026.05 | 10 | |
| SAVGOTraining steps=1,000,000, Seeds=102026.05 | 10 |