Download now Free registration required
Machine learning has become a fundamental tool for computer security since it can rapidly evolve to changing and complex situations. That adaptability is also vulnerability: attackers can exploit machine learning systems. This paper presents a taxonomy identifying and analyzing attacks against machine learning systems. The authors show how these classes influence the costs for the attacker and defender, and the authors give a formal structure defining their interaction.
- Format: PDF
- Size: 215.5 KB