Policy Evaluation on 2x2 Queueing Network rho = (0.90, 0.90)
1,316.9ValueBTSD-PPO
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| BTSD-PPOc^ov=0.62026.06 | 1,316.9 | 2.1 | 2.5 | |
| Exact optimalc^ov=0.62026.06 | 1,285.3 | 2.1 | — | |
| BTSD-PPOc^ov=0.32026.06 | 1,261 | 2.2 | 1.6 | |
| Exact optimalc^ov=0.32026.06 | 1,241.1 | 2 | — | |
| BTSD-PPOc^ov=0.12026.06 | 1,226.3 | 1.8 | 1.3 | |
| Exact optimalc^ov=0.12026.06 | 1,210.4 | 1.8 | — |