Policy Evaluation on 2x2 Queueing Network rho = (0.95, 0.85)
1,185.4Expected ValueExact optimal
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Exact optimalc^ov=0.12026.06 | 1,185.4 | 2 | — | |
| BTSD-PPOc^ov=0.12026.06 | 1,198.1 | 1.6 | 1.1 | |
| Exact optimalc^ov=0.32026.06 | 1,218.1 | 2 | — | |
| BTSD-PPOc^ov=0.32026.06 | 1,234.2 | 1.9 | 1.3 | |
| Exact optimalc^ov=0.62026.06 | 1,264.5 | 2 | — | |
| BTSD-PPOc^ov=0.62026.06 | 1,278.2 | 1.9 | 1.1 |