Evaluating AI Agents for Cyber Defense: A Comparison of Deep Reinforcement Learning and LLM Approaches

This paper presents a comparative evaluation of Deep Reinforcement Learning (DRL) and Large Language Model (LLM) based approaches for autonomous cyber defense. We assess both paradigms across simulated cybersecurity environments, measuring intrusion detection accuracy, response time, and adaptability to novel attack patterns.

Authors: Hassan Chowdhry, J. Manero, and S. Sampalli

Venue: International Conference on Intelligent Data Engineering and Automated Learning (IDEAL 2025)

Recommended citation: H. Chowdhry, J. Manero, and S. Sampalli. (2025). "Evaluating AI Agents for Cyber Defense: A Comparison of Deep Reinforcement Learning and LLM Approaches." IDEAL 2025.
Download Paper