/CSE 539. AI Robustness and Privacy

Catalog description

This course systematically introduces the foundational concepts of Trustworthy AI, emphasizing the interpretability, general robustness, adversarial robustness (adversarial attacks and defenses), poisoning attacks and defenses, backdoor robustness (backdoor attacks and defenses), privacy (data leakage and model stealing), differential privacy, federated learning, fairness, data. Students will explore the multifaceted dimensions of Trustworthy AI and gain practical experience in designing, implementing, and evaluating AI systems that align with ethical standards and societal values. By taking this course, students will gain a comprehensive understanding of trustworthy AI and independently complete hands-on projects and a self-selected research topic. Prerequisite: CYB 134 , CSE 274 or CSE 232 .