AutoSND: From Execution Evidence to Structural Policies for Automated Network Dismantling Heuristic Discovery

2026-08-04

AI Summary

This work addresses the trade-off between effectiveness and efficiency in network dismantling, a challenge inadequately met by existing methods that either rely on handcrafted heuristics or struggle to extract actionable strategies from execution trajectories. To bridge this gap, the authors propose AutoSND, a three-stage tree search framework: it first performs broad exploration using simple heuristics while recording execution evidence; then induces interpretable structural policies concerning local signals, neighborhood visitation, and state-update scope; and finally leverages these policies to guide an efficient search for high-quality solutions. AutoSND is the first method to systematically transform execution evidence into explainable, structured heuristics, marking a shift from black-box generation to structural discovery. Experiments on 12 real-world and 3 large-scale networks demonstrate that AutoSND-Q/S outperforms baselines in both performance and stability, with its learned heuristics centering on residual degree, bounded local signal adjustment for node ordering, and constrained state-update ranges.

Abstract

Network dismantling is fundamental to analyzing the robustness and vulnerability of complex systems, yet practical heuristics must balance effectiveness and computational efficiency, and are usually designed manually by researchers. Existing large language model based automatic heuristic design methods can generate and screen candidates, yet they have difficulty further transforming candidate quality or failure states during execution into structural-level guid- ance for subsequent generation. We propose AutoSND, a three stage tree search framework for complete network dismantling pro- grams. Stage I broadly explores from simple heuristics and archives execution evidence. Stage II compiles candidate records into struc- tural policies concerning local signals, neighborhood access, and state update ranges. Stage III continues tree search conditioned on these policies and obtains the final quality prioritized and speed prioritized candidates, AutoSND-Q/S. Experiments on 12 real world networks and 3 large real world networks show that AutoSND achieves better search performance and stability and discovers more competitive and structurally interpretable network disman- tling programs. The final candidates form an interpretable structure that uses residual degree as the backbone, adjusts node order with bounded local signals, and restricts the state update range. Code is available at https://github.com/MirrorNew/AutoSND.

Problem

• network dismantling
• heuristic discovery
• structural policies
• execution evidence
• automated design

Innovation

• automated heuristic discovery
• network dismantling
• structural policy learning
• tree search framework
• execution evidence