Online Anomaly Detection under Adversarial Impact
Marius Kloft and Pavel Laskov
In: JMLR Workshop and Conference Proceedings 9 (AISTATS 2010)., 12 May -- 14 May 2010, Sardinia, Italy.

Abstract

Security analysis of learning algorithms is gaining increasing importance, especially since they have become target of deliberate obstruction in certain applications. Some security-hardened algorithms have been pre- viously proposed for supervised learning; however, very little is known about the be- havior of anomaly detection methods in such scenarios. In this contribution, we analyze the performance of a particular method| online centroid anomaly detection|in the presence of adversarial noise. Our analysis addresses three key security-related issues: derivation of an optimal attack, analysis of its eciency and constraints. Experimental evaluation carried out on real HTTP and ex- ploit traces conrms the tightness of our the- oretical bounds.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.

[Edit]