[论文] Robustness of Anomaly Detection Models for Industrial Control Systems ...

论文概要

研究领域: ML 作者: Mustafa Umut Ozbek, Taiwo Ojo, Pooria Madani 发布时间: 2025-08-26 arXiv: 2508.17623

中文摘要

基于机器学习的异常检测在工业控制系统（ICS）中应用日益广泛，但大多数研究假设检测器训练数据是可信赖的。实际上，训练数据可能通过受损日志、标注错误、篡改的历史记录或不安全的重训练过程而被污染。本文在SWaT基准数据集上评估了离线ICS异常检测管线的训练时数据污染鲁棒性。我们评估了11种异构异常检测器在三种污染策略下的表现：随机注入、相似性目标注入和特征噪声注入。前两种将攻击样本插入正常训练池，第三种向选定的正常训练样本添加有界高斯噪声。这些攻击属于污染型而非梯度驱动的投毒方法。在1%到10%的污染预算下，使用干净的验证和测试集进行评估。结果表明鲁棒性高度依赖模型，且无法仅从干净数据性能预测。注入型污染造成最大退化，尤其对局部密度和距离型检测器，而特征噪声污染影响相对有限。PCA、SVM、HBOS和IForest相对稳定，调优后的神经网络检测器表现出中等鲁棒性。

原文摘要

Machine-learning-based anomaly detection is increasingly used in industrial control systems (ICS), yet most studies assume that detector training data is trustworthy. In practice, training data may be corrupted through compromised logs, labeling errors, manipulated historian records, or unsafe retraining processes. This paper evaluates the robustness of offline ICS anomaly-detection pipelines on the Secure Water Treatment (SWaT) benchmark under training-time contamination. We assess 11 heterogeneous anomaly detectors under three contamination strategies: random injection, similarity-targeted injection, and feature-noise injection. The first two insert attack samples into the nominal training pool, while the third adds bounded Gaussian noise to selected normal training samples. These attack...

--- *自动采集于 2026-08-26*

#论文 #arXiv #ML #小凯