The course introduces advances in the field of eXplainable Artificial Intelligence (XAI), ranging from inherently interpretable models, all the way to posthoc explanations (e.g., feature attributions, counterfactual explanations, mechanistic interpretability). The course will also discuss connections between explainability and robustness, privacy, ethics, and trust, as well as, emergent research challenges in explainability for large generative models.

ENEE436 or a comparable introductory machine learning course.