Statistical Learning and Data Analytics

A graduate-level treatment of machine learning that frames every model — from ordinary least squares up through deep nets and graphical models — as a statistical inference problem, so you learn when to reach for Bayesian versus frequentist tools and how to actually reason about generalization rather than just call a library. Expect written midterm and final exams, four quizzes that keep the probability and linear algebra sharp, a term project where you implement a full pipeline end to end, and a research essay surveying a sub-area in depth. It assumes you are comfortable with probability and linear regression at the undergraduate level and serves as the methodological backbone for later work in computer vision, signal processing, or any research group doing data-driven modeling.