An Explainable Generic Design of Self-Evolving Intelligent Security Systems for Cyber-Attacks Detection

Over the last few years, cyber-attacks have posed significant and persistent challenges to governments, businesses, and individuals, driven by the increasing reliance on digital technologies. Intrusion Detection Systems (IDSs) are among the most effective solutions against cyber threats and have been a major focus of cyber security research.

In this seminar, we will present our recent contributions to the design of intelligent IDSs that prioritize transparency and explainability. First, we will discuss the limitations of existing IDSs, particularly their applicability to large-scale networks connecting diverse devices. Next, we will present our solutions, highlighting the rationale behind each design choice. Finally, we will discuss the remaining limitations of our approaches and potential directions for future improvement.