Exploring the Efficacy of Multi-Agent Reinforcement Learning for Autonomous Cyber Defence: A CAGE Challenge 4 Perspective

作者：Mitchell Kiely, Metin Ahiskali, Étienne Borde, Benjamin Bowman, David Bowman, Dirk Van Bruggen, K H Cowan, Prithviraj Dasgupta, Erich Devendorf, Ben Edwards, Alex Fitts, Sunny Fugate, Ryan Gabrys, W. T. S. Gould, Hao Huang, Jules Jacobs, Ryan Kerr, Isaiah J. King, Li Li, Luis Martínez‐Sobrido, Christopher R. Moir, Craig E. Murphy, Olivia Naish, Claire Owens, Miranda Purchase, Ahmad Ridley, Adrian Taylor, Sarah Farmer, William Valentine, Yiyi Zhang · 发表于：Proceedings of the AAAI Conference on Artificial Intelligence · 年份：2025 · DOI：10.1609/aaai.v39i28.35158 · 被引用次数：14 · 研究领域：Advanced Malware Detection Techniques、Network Security and Intrusion Detection、Smart Grid Security and Resilience

As cyber threats become increasingly automated and sophisticated, novel solutions must be introduced to improve defence of enterprise networks. Deep Reinforcement Learning (DRL) has demonstrated potential in mitigating these advanced threats. Single DRL Agents have proven utility toward execution of autonomous cyber defence. Despite the success of employing single DRL Agents, this approach presents significant limitations, especially regarding scalability within large enterprise networks. An attractive alternative to the single agent approach is the use of Multi-Agent Reinforcement Learning (MARL). However, developing MARL agents is costly with few options for examining MARL cyber defence techniques against adversarial agents. This paper presents a MARL network security environment, the fourth iteration of the Cyber Autonomy Gym for Experimentation (CAGE) challenges. This challenge was specifically designed to test the efficacy of MARL algorithms in an enterprise network. Our work aims to evaluate the potential of MARL as a robust and scalable solution for autonomous network defence.