Humanoid Control on MuJoCo (Abrupt-change recovery time)
7.5Recovery Time (%)MEow+AES
Evaluation Results
| Method | Links | |
|---|---|---|
| MEow+AESBase RL Algorithm=MEow, Entropy Scheduling (AES)=Enabled2026.01 | 7.5 | |
| SAC+AESBase RL Algorithm=SAC, Entropy Scheduling (AES)=Enabled2026.01 | 8.6 | |
| PPO+AESBase RL Algorithm=PPO, Entropy Scheduling (AES)=Enabled2026.01 | 10.3 | |
| SQL+AESBase RL Algorithm=SQL, Entropy Scheduling (AES)=Enabled2026.01 | 10.6 | |
| PPOBase RL Algorithm=PPO, Entropy Scheduling (AES)=Disabled2026.01 | 13 | |
| SACBase RL Algorithm=SAC, Entropy Scheduling (AES)=Disabled2026.01 | 14.8 | |
| MEowBase RL Algorithm=MEow, Entropy Scheduling (AES)=Disabled2026.01 | 15.1 | |
| SQLBase RL Algorithm=SQL, Entropy Scheduling (AES)=Disabled2026.01 | 16.3 |