Probabilistic Machine Learning and Statistical Modeling

3 credits · Graduate School

This course introduces the foundational principles of probabilistic machine learning and modern statistical modeling, emphasizing the mathematical structure underlying common learning algorithms. Topics include discriminant analysis, kernel methods, tree-based methods, neural networks, dimensionality reduction, and clustering. Students will develop a unified probabilistic perspective on machine learning grounded in probability theory, statistical inference, decision theory, information theory, and optimization. The course balances theoretical understanding with practical modeling insight, preparing students to critically evaluate and apply machine learning methods in biomedical and data science contexts. Assignments focus on mathematical reasoning, conceptual clarity, and interpretation of modeling assumptions. Biomedical case studies are used to illustrate real-world deployment of statistical learning algorithms. Prerequisites: BIOS 6342. [3]