让防御者在对抗中更智能地加固网络，避免被攻击者钻空子。
Decision-Focused Learning in Network Interdiction Games
- 用对抗性场景替代普通数据训练，解决原有方法的决策失效问题。
- 实验证明新方法在真实和合成网络中均能恢复端到端优化优势。
- 适合研究网络防御、博弈学习与鲁棒优化的学者参考。
我们研究最短路径网络干扰（SPNI）博弈中的决策聚焦学习（DFL），这是一种斯塔克尔伯格博弈：干扰者（领导者）强化网络弧以抵御攻击，而逃避者（追随者）对攻击成本不确定，依赖机器学习预测器选择最短路径。尽管DFL作为端到端优化框架效果显著，但我们在该博弈设定下发现其存在根本性结构缺陷：其训练目标存在广泛的决策等价类成本估计器，虽在名义上损失为零，但在实际干扰下却完全失效，逆转了其相对于传统预测聚焦学习（PFL）的优势。为此，我们提出对抗性决策聚焦学习（A-DFL），通过用干扰后的场景替换名义训练样本，消除有害的等价类。在合成与真实网络上的实验表明，A-DFL成功恢复了该设定下DFL的优化优势，实现了有效的端到端优化。
原文摘要 · Abstract (English)
We study decision-focused learning (DFL) in shortest-path network interdiction (SPNI) games, a Stackelberg game where an interdictor (leader) strengthens the networks' arcs against attacks, while an evader (follower) who is uncertain about costs of attacking network arcs relies on a machine-learned predictor to identify the shortest path. While DFL is highly effective as an end-to-end optimization framework, we show that it faces a fundamental structural failure when employed in this game setting: its training objective admits a broad decision-equivalence class of cost estimators that achieve zero nominal loss yet fail under interdiction, reversing DFL's usual advantage over a naive prediction-focused learning (PFL) approach. To address this, we propose Adversarial DFL (A-DFL), which replaces nominal training samples with interdicted scenarios to collapse the harmful equivalence class. Experiments on synthetic and real-world networks confirm that A-DFL restores DFL's advantage in this game setting, enabling effective end-to-end optimization.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。