PPO policy for grid topology control is distilled into decision trees and random forests that outperform the teacher on reward and survival time with lower inference cost and high interpretability.
arXiv preprint arXiv:2003.07339 (2020)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Interpretable Policy Distillation for Power Grid Topology Control
PPO policy for grid topology control is distilled into decision trees and random forests that outperform the teacher on reward and survival time with lower inference cost and high interpretability.