Continuous Control on HalfCheetah v5 (Final return)
11,931.2Final ReturnSDAC
Evaluation Results
| Method | Links | |
|---|---|---|
| SDACNFE=20 × 322026.04 | 11,931.2 | |
| SACNFE=12026.04 | 11,349.7 | |
| TRFP(ours)NFE=4 × 42026.04 | 10,867.5 | |
| TRFP(one-step)NFE=1 × 42026.04 | 10,332.7 | |
| TD3NFE=12026.04 | 8,677.1 | |
| MaxEntDPNFE=20 × 102026.04 | 6,794.8 |