Welcome to CSE 291H  Statistical learning and combinatorics Time: MW 3:305:00 OverviewThe class will focus on two themes: combinatorics and learning theory. We will cover equivalences between basic notions in statistical learning theory and combinatorial objects, and use these equivalences for studying combinatorics, statistics, machine learning, and geometry. Target audience: The course is aimed at graduate students in machine learning and theoretical computer science. Mathematical maturity is expected and familiarity with fundamental concepts in linear algebra, analysis, probability, and discrete mathematics is required (read the preassignment for more information). Grading: This is a 4 unit course. The assesement in this course will be based on homework (70%), and preparing scribe notes in latex (30%). Homework can be done individually or in pairs. Please write your solutions in latex/word, save as pdf, and email to Shay by the due date. Tentative plan: We will begin with combinatorial topics related to VC dimension and other quantities related to learning theory (including advanced topics). Then proceed to the equivalence between VC dimension and uniform convergence, and between compression and learnability. Exercises
