Reinforcement Learning on Minepump 1.0 (train)
-806.73RewardDQN + Static (*)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DQN + Static (*)Shielding=Static (*)2025.11 | -806.73 | 100 | |
| DQN + AdaptiveShielding=Adaptive2025.11 | -807.21 | 100 | |
| DQN + Static 2Shielding=Static 22025.11 | -816.33 | 0 | |
| Static Ctrl. 2Type=Symbolic Controller2025.11 | -827.87 | 100 | |
| Static Ctrl. 1Type=Symbolic Controller2025.11 | -849.64 | 100 | |
| DQN + Static 1Shielding=Static 12025.11 | -882.25 | 0 | |
| DQNShielding=None2025.11 | -883.1 | 0 |