Navigation on D4RL AntMaze umaze-diverse v2
56.6Initial Normalized D4RL ScoreIQL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| IQLAlgorithm=IQL, Evaluation Interactions=200k2024.12 | 56.6 | 35.8 | |
| O2PPOAlgorithm=O2PPO, Evaluation Interactions=200k2024.12 | 56.4 | 86.3 | |
| O2SACAlgorithm=O2SAC, Evaluation Interactions=200k2024.12 | 43.8 | 79.8 | |
| O2TD3Algorithm=O2TD3, Evaluation Interactions=200k2024.12 | 38.4 | 52.2 | |
| Cal-QLAlgorithm=Cal-QL, Evaluation Interactions=200k2024.12 | 23.8 | 71.2 | |
| PEXAlgorithm=PEX, Evaluation Interactions=200k2024.12 | 12.6 | 16 |