Adversarial Attacks and Defense Mechanisms in the Age of Quantum Computing
Yara Shamoo · Advances in information security, privacy, and ethics book series · 2024
Adversarial attacks on machine learning models have become a significant concern in cybersecurity, especially with the advent of quantum computing. These attacks aim to manipulate the decision-making process of AI systems, leading to vulnerabilities that can be exploited by malicious actors. As quantum computing promises to revolutionize various industries, it also introduces new challenges for defending against adversarial threats. This chapter explores the impact of quantum computing on adversarial machine learning, examining how quantum algorithms can be both a tool for enhancing attack strategies and a foundation for developing more robust defense mechanisms. It reviews existing defense techniques, such as adversarial training and gradient masking, and discusses the potential for quantum-aware models to counteract these threats.