analiza aiqbench
This research critically evaluates the vulnerability of machine learning-based anomaly detection systems in industrial control environments when training data is compromised. The study demonstrates that robustness varies significantly by model architecture, with injection-based attacks causing severe degradation in density-based detectors while tree ensembles and PCA-based methods remain relatively stable. These findings underscore the necessity of validating training data integrity rather than relying solely on clean-data performance metrics for safety-critical applications.
Robustness of Anomaly Detection Models for Industrial Control Systems under Training-Time Data Contamination
arXiv:2608.23547v1 Announce Type: cross Abstract: 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 attacks are contamination-based rather than gradient-driven poisoning methods. Contamination budgets from 1% to 10% are evaluated using clean validation and test sets under a unified offline protocol. The results show that robustness is strongly model-dependent and cannot be predicted from clean-data performance alone. Injection-based contamination causes the greatest degradation, particularly for local-density and distance-based detectors, whereas feature-noise contamination has a comparatively limited effect. PCA, SVM, HBOS, and IForest remain relatively stable, while the tuned neural detectors demonstrate intermediate robustness. Overall, the findings highlight the importance of training-data integrity in ML-enabled ICS monitoring, subject to the evaluated dataset, models, and threat assumptions.
Abstract: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 attacks are contamination-based rather than gradient-driven poisoning methods. Contamination budgets from 1% to 10% are evaluated using clean validation and test sets under a unified offline protocol. The results show that robustness is strongly model-dependent and cannot be predicted from clean-data performance alone. Injection-based contamination causes the greatest degradation, particularly for local-density and distance-based detectors, whereas feature-noise contamination has a comparatively limited effect. PCA, SVM, HBOS, and IForest remain relatively stable, while the tuned neural detectors demonstrate intermediate robustness. Overall, the findings highlight the importance of training-data integrity in ML-enabled ICS monitoring, subject to the evaluated dataset, models, and threat assumptions.
Accepted and to appear in IEEE CASCON 2026. Code is available at: this https URL
Cryptography and Security (cs.CR); Machine Learning (cs.LG)
arXiv:2608.23547 [cs.CR]
(or arXiv:2608.23547v1 [cs.CR] for this version)
https://doi.org/10.48550/arXiv.2608.23547 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 315 cuvinte
Subiecte conexe
Article: Governing AI in the Cloud: A Practical Guide for Architects
InfoQ · 15 iun. 2026
Article: Artificial Intelligence-Driven Phishing: How Phishing Technique Is Evolving and Implemented
InfoQ · 8 iun. 2026
AI Swiftly Detects and Neutralises 6G Network Threats in Milliseconds
Quantum Zeitgeist · 10 iun. 2026
SysML v2 Drives Digital Twin Framework for Quantum Networks
Quantum Zeitgeist · 14 iul. 2026