The security of machine learning

TLDR

Machine learning’s rapid adaptability makes it a key tool for computer security while simultaneously exposing it to exploitation by attackers. The authors present a taxonomy that identifies and analyzes attacks against machine learning systems. They develop a formal framework that defines attack classes, their cost implications, and use it to survey the literature and illustrate guidance for attacks on SpamBayes. The taxonomy suggests new lines of defense against such attacks.

Abstract

Machine learning’s ability to rapidly evolve to changing and complex situations has helped it become a fundamental tool for computer security. That adaptability is also a vulnerability: attackers can exploit machine learning systems. We present a taxonomy identifying and analyzing attacks against machine learning systems. We show how these classes influence the costs for the attacker and defender, and we give a formal structure defining their interaction. We use our framework to survey and analyze the literature of attacks against machine learning systems. We also illustrate our taxonomy by showing how it can guide attacks against SpamBayes, a popular statistical spam filter. Finally, we discuss how our taxonomy suggests new lines of defenses.