Continuous Control on Ant v5 (Final Return)
4,802.1Final ReturnTRFP(ours)
Evaluation Results
| Method | Links | |
|---|---|---|
| TRFP(ours)NFE=4 × 42026.04 | 4,802.1 | |
| TRFP(one-step)NFE=1 × 42026.04 | 4,792.8 | |
| TD3NFE=12026.04 | 4,708.5 | |
| SACNFE=12026.04 | 4,566.4 | |
| MaxEntDPNFE=20 × 102026.04 | 2,888.6 | |
| SDACNFE=20 × 322026.04 | 977.5 |