Continuous Control on MuJoCo HumanoidStandup standard (test)
146,529.7Episodic ReturnRAT
Evaluation Results
| Method | Links | |
|---|---|---|
| RATArchitecture=Shared actor-critic networks, Seeds=5 random seeds2026.05 | 146,529.7 | |
| PPOArchitecture=Shared actor-critic networks, Seeds=5 random seeds2026.05 | 130,014.2 | |
| ACKTRArchitecture=Shared actor-critic networks, Seeds=5 random seeds2026.05 | 127,928.5 | |
| SophiaArchitecture=Shared actor-critic networks, Seeds=5 random seeds2026.05 | 111,212.6 |