Topology Control on case36 IEEE modified (evaluation)
158,549.3Average RewardGreedy oracle
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Greedy oracleDecision time (ms/step)=184.1992026.04 | 158,549.3 | 1,866.8 | |
| Physics-Informed Reinforcement Learning with Gibbs PriorsDecision time (ms/step)=0.9242026.04 | 149,963.1 | 1,774.2 | |
| PPODecision time (ms/step)=4.0812026.04 | 120,540.6 | 1,390.8 | |
| RandomDecision time (ms/step)=0.0052026.04 | 1,138.1 | 42.6 |