Trajectory Ranking on LunarLander v2
207.13Average RewardV-Goal
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| V-Goal2025.12 | 207.13 | 319.2 | |
| Naive Normalization2025.12 | 188.12 | 433.2 | |
| Entropy-Based2025.12 | 121.27 | 871 | |
| V-Normalization2025.12 | 120.59 | 1,000 | |
| Bellman Error2025.12 | 117.37 | 1,000 | |
| Classic (ΔQ only)2025.12 | 116.87 | 1,000 |