Assessing the Applicability of Adversarial Machine Learning Approaches for Cybersecurity
Prashamsh Takkalapally, Nandan Sharma, Arjun Jaggi, Karim Hudani, Ketan Prabhunath Gupta, Yuvaraj Natarajan · 2024
Adverse machine learning (AML) is a rising area of study that uses system-mastering algorithms to perceive malicious hobbies or to discover malicious adversaries in cybersecurity settings. This research domain combines device-gaining knowledge with game principles to assess and expect the behavior of adversaries. This paper discusses the capacity of using AML tactics for cybersecurity and provides an overview of existing research study findings and safety literature to help the dialogue. Specifically, the paper investigates the security challenges posed by using AML techniques, together with the technical regulations and pointers, to ensure their relaxed and effective deployment. Moreover, the paper offers recommendations and future instructions for furthering studies on AML in cybersecurity. Typical, the evaluation highlights the significance of assessing the applicability of AML techniques for security applications, in addition to its capability to enhance the effectiveness of security systems.