Continuous Control on MuJoCo Hopper standard (test)
2,346.8Episodic ReturnPPO
Evaluation Results
| Method | Links | |
|---|---|---|
| PPOArchitecture=Shared actor-critic networks, Seeds=5 random seeds2026.05 | 2,346.8 | |
| RATArchitecture=Shared actor-critic networks, Seeds=5 random seeds2026.05 | 2,334.6 | |
| ACKTRArchitecture=Shared actor-critic networks, Seeds=5 random seeds2026.05 | 2,138.9 | |
| SophiaArchitecture=Shared actor-critic networks, Seeds=5 random seeds2026.05 | 1,104 |