Interpretability and Explainability in Machine Learning

4 credits

This course will cover advances in interpretable and explainable machine learning. Topics include modern methods to learn simple, transparent models (e.g., rule lists, decision trees, sparse linear models); methods for post-hoc explainability (e.g., SHAP, LIME, saliency maps, prototype-generation methods); and overarching concepts in the interpretation, description, explanation of machine learning (e.g., model multiplicity, cognitive biases, user studies). Prerequisites: DSC 240. Students should be comfortable reading research papers with a critical perspective; have prior course work in machine learning (e.g., DSC 240); and have some interest in interpretability and explainability for your research or career. Restricted to major codes DSC-MS-001 and DSC-PhD-001.