Explainable AI for Cybersecurity

This tutorial will provide a comprehensive overview of security attacks as well as detection techniques using explainable AI. Specifically, the tutorial will consist of six parts. The first part will outline a wide variety of software and hardware security threats and vulnerabilities. The second part will cover various machine learning algorithms, including decision tree, random forest, deep neural network, recurrent neural network, unsupervised learning, zero-shot learning, and reinforcement learning. The third part will introduce explainable AI algorithms to interpret machine learning modelsí behaviors in a human-understandable way, using model distillation, Shapley value analysis, and integrated gradients. The fourth part will discuss state-of-the-art attack detection using explainable AI. The fifth part will cover how to enable hardware acceleration of explainable AI models for real-time vulnerability detection. Finally, it will discuss the security threats toward machine learning models (adversarial attack, poisoning attack, and AI Trojan attack), and effective countermeasures to design robust AI models.