A Study on the Efficacy of Deep Reinforcement Learning for Intrusion Detection

The world has witnessed a fast-paced digital transformation in the past decade, giving rise to all-connected environments. While increasingly widespread availability of networks benefited many aspects our lives, providing necessary infrastructure for smart autonomous systems, it also created large cyber attack surface. This made real-time network intrusion detection significant component any computerized system. With advances computer hardware architectures with fast, high-volume data processing capabilities and developments field artificial intelligence, deep learning emerged as aid achieving accurate detection, especially zero-day attacks. In this paper, we propose reinforcement learning-based approach demonstrate its efficacy using two publicly available datasets, namely NSL-KDD UNSW-NB15. experiment results suggest that potential provide effective complex future.