Practical Attacks on Machine Learning: A Case Study on Adversarial Windows Malware

TL;DR

This paper explores how to systematically evaluate the security of machine learning models against adversarial attacks, using a case study on Windows malware detection to demonstrate practical attack methods.

Contribution

It introduces a framework for automated, scalable security evaluation of machine learning systems through practical adversarial attacks, exemplified by a Windows malware detection case study.

Findings

Demonstrated practical adversarial attacks on malware detection models

Provided a scalable framework for security evaluation of ML systems

Highlighted vulnerabilities in machine learning-based security tools

Abstract

While machine learning is vulnerable to adversarial examples, it still lacks systematic procedures and tools for evaluating its security in different application contexts. In this article, we discuss how to develop automated and scalable security evaluations of machine learning using practical attacks, reporting a use case on Windows malware detection.