2D bipedal locomotion on Hardcore (OpenAI Gym) (test)
140.1Average ReturnADD
Evaluation Results
| Method | Links | |
|---|---|---|
| ADDTraining Algorithm=PPO, Environmental Steps=2 billion, Guidance=true2024.10 | 140.1 | |
| PLRTraining Algorithm=PPO, Environmental Steps=2 billion2024.10 | 116.6 | |
| ACCELTraining Algorithm=PPO, Environmental Steps=2 billion2024.10 | 53.1 | |
| ADDTraining Algorithm=PPO, Environmental Steps=2 billion, Guidance=false2024.10 | 35.5 | |
| DRTraining Algorithm=PPO, Environmental Steps=2 billion2024.10 | 10.6 | |
| PAIREDTraining Algorithm=PPO, Environmental Steps=2 billion2024.10 | -47.2 |