Continuous Control on MuJoCo Ant v4 (test)
6,665Mean Episodic ReturnFP-DRL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| FP-DRL2026.04 | 6,665 | 0 | |
| DACER2026.04 | 5,971 | 410 | |
| SAC-FLOW2026.04 | 5,312 | 1,622 | |
| SAC2026.04 | 5,179 | 1,408 | |
| DSAC-T2026.04 | 3,499 | 1,081 | |
| TD32026.04 | 3,023 | 0 | |
| RATArchitecture=Shared actor-critic networks, Seeds=5 random seeds2026.05 | 2,926.6 | — | |
| PPOArchitecture=Shared actor-critic networks, Seeds=5 random seeds2026.05 | 1,373.9 | — | |
| ACKTRArchitecture=Shared actor-critic networks, Seeds=5 random seeds2026.05 | 23.4 | — | |
| SophiaArchitecture=Shared actor-critic networks, Seeds=5 random seeds2026.05 | -7 | — |