Adversarial classification: An adversarial risk analysis approach

TL;DR

This paper introduces an adversarial risk analysis framework for security classification problems, offering an alternative to game-theoretic methods that rely on unrealistic assumptions, and discusses computational aspects.

Contribution

It proposes a novel adversarial risk analysis approach for security classification, addressing limitations of existing game-theoretic methods.

Findings

Demonstrates the effectiveness of the adversarial risk analysis framework through examples

Highlights computational and implementation considerations

Provides a practical alternative to traditional game-theoretic approaches

Abstract

Classification problems in security settings are usually contemplated as confrontations in which one or more adversaries try to fool a classifier to obtain a benefit. Most approaches to such adversarial classification problems have focused on game theoretical ideas with strong underlying common knowledge assumptions, which are actually not realistic in security domains. We provide an alternative framework to such problem based on adversarial risk analysis, which we illustrate with several examples. Computational and implementation issues are discussed.