Navigation on D4RL AntMaze medium-diverse v2
85.5Normalized D4RL ScoreO2PPO
Evaluation Results
| Method | Links | |
|---|---|---|
| O2PPOLearning Stage=Online fine-tuning, Environment steps=250,0002024.12 | 85.5 | |
| PEXLearning Stage=Online fine-tuning2024.12 | 83.4 | |
| IQLLearning Stage=Online fine-tuning2024.12 | 73.2 | |
| PEXLearning Stage=Offline2024.12 | 70.8 | |
| O2SACLearning Stage=Online fine-tuning, Environment steps=200,000, Double Q networks=false2024.12 | 70.6 | |
| IQLLearning Stage=Offline2024.12 | 68.8 | |
| O2PPOLearning Stage=Offline2024.12 | 63.5 | |
| O2SACLearning Stage=Offline2024.12 | 61.6 | |
| AWACLearning Stage=Offline2024.12 | 0 | |
| AWACLearning Stage=Online fine-tuning2024.12 | 0 |