Offline Multi-Agent Reinforcement Learning on Marine-Easy Medium-Replay
87.5Performance (3m)HiSSD
Evaluation results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HiSSDTime steps=10K2026.03 | 87.5 | 85 | 85.6 | 66.9 | 100 | 99.4 | 96.9 | 80.6 | 53.1 | 36.3 | 79.1 | |
| STAIRS-FormerTime steps=10K2026.03 | 86.9 | 89.4 | 56.9 | 79.4 | 91.3 | 90.6 | 83.1 | 82.5 | 55 | 49.4 | 76.5 | |
| ODISTime steps=10K2026.03 | 50 | 0 | 0 | 15.6 | 0 | 0 | 0 | 0 | 0 | 0 | 6.6 | |
| UPDeT-mTime steps=10K2026.03 | 25.8 | 0 | 0 | 0 | 0 | 0 | 0.8 | 0 | 0 | 0 | 2.7 |