A Deep Reinforcement Learning Approach for Anomaly Network Intrusion Detection System

Network intrusion detection systems (NIDS) are essential for organizations to ensure the safety and security of their communication and information. In this paper, we propose a deep reinforcement learning-based (DRL) for anomaly network intrusion detection system. Our system has the ability of self-updating to reflect new types of network traffic behavior. This study includes three major contributions. First, to show the overall applicability of our approach, we demonstrate our work through two well-known NIDS benchmark datasets: NSL-KDD and UNSW-NB15 and a real campus network log. Second, to show the feasibility of our approach, we compared our method with three other classic machine learning methods and two related published results. Third, our model is capable of processing a million scale of network traffic on a real-time basis.