Intrusion Detection Systems (IDS) play a crucial role in computer network security by identifying malicious activities and potential cyber attacks. In the past, traditional machine learning methods have been widely used for intrusion detection; however, there are challenges in adapting to new and emerging attack types. This thesis combines machine learning and cybersecurity by applying Deep Reinforcement Learning (DRL) in intrusion detection using the NSL-KDD dataset. We design and implement an DRL-based framework in which an agent learns to classify network traffic by interacting with an environment and receives rewards based on detection accuracy. We also look at the importance of feature selection and classification techniques and how effective they are in improving detection performance, reducing computational complexity, and enhancing model interpretability. This study highlights the potential of reinforcement learning as a cutting-edge approach to improve modern intrusion detection systems.
