Navigation on D4RL AntMaze large-diverse v2
58.4Normalized D4RL ScorePEX
Evaluation Results
| Method | Links | |
|---|---|---|
| PEXLearning Stage=Online fine-tuning2024.12 | 58.4 | |
| O2PPOLearning Stage=Online fine-tuning, Environment steps=250,0002024.12 | 50.33 | |
| O2SACLearning Stage=Online fine-tuning, Environment steps=200,000, Double Q networks=false2024.12 | 40.2 | |
| PEXLearning Stage=Offline2024.12 | 40 | |
| IQLLearning Stage=Online fine-tuning2024.12 | 36 | |
| IQLLearning Stage=Offline2024.12 | 31 | |
| O2PPOLearning Stage=Offline2024.12 | 26.83 | |
| O2SACLearning Stage=Offline2024.12 | 24.2 | |
| AWACLearning Stage=Offline2024.12 | 0 | |
| AWACLearning Stage=Online fine-tuning2024.12 | 0 |