Enhancing Cybersecurity Protocols with Reinforcement Learning

EasyChair Preprint 14340

Abstract

In the evolving landscape of cybersecurity, traditional protocols often struggle to keep pace with sophisticated and dynamic threats. This paper explores the integration of Reinforcement Learning (RL) techniques to enhance cybersecurity protocols. Reinforcement Learning, a type of machine learning where an agent learns to make decisions by receiving rewards or penalties, offers a promising approach for developing adaptive and autonomous security systems. By modeling cybersecurity challenges as RL problems, this approach enables protocols to learn from interactions with their environment, continuously improving their ability to detect and respond to threats. The paper reviews current methodologies in applying RL to various aspects of cybersecurity, including intrusion detection, threat response, and vulnerability management. It also discusses the potential benefits, such as increased adaptability and efficiency, as well as challenges, including computational requirements and the need for robust training environments. The study aims to provide insights into how RL can be leveraged to build more resilient cybersecurity frameworks and proposes directions for future research in this emerging intersection of AI and security.

Keyphrases: Cyber Security, learning, machine

@booklet{EasyChair:14340,
  author    = {Obaloluwa Ogundairo and Peter Broklyn},
  title     = {Enhancing Cybersecurity Protocols with Reinforcement Learning},
  howpublished = {EasyChair Preprint 14340},
  year      = {EasyChair, 2024}}