A Machine Learning-Based Survey Of Adversarial Attacks And Defenses In Malware Classification

As the number and complexity of malware attacks continue to increase, there is a growing need for effective malware detection methods. Machine learning models are effective at Malware detection, but show some vulnerabilities after Adversarial Attacks. Adversarial attacks can be used to generate malicious files that evade detection by these models, which can be a serious cybersecurity threat. This survey article presents an in-depth examination of recent research in adversarial attack and defense approaches for malware classification in cybersecurity. It categorizes the methods into four main categories: generative models, feature-based methods, ensemble methods, and hybrid methods. The paper presents the datasets and evaluation metrics commonly used in this field. At last, it identifies open research challenges and provides a direction for future research purpose. This paper serves as a valuable resource for researchers and practitioners working on malware classification and cyber security.