Non-parametric Statistical Learning

Catalog description

This course is an introduction to two important modern fields of statistics: nonparametric statistics and modern statistical learning theory, the theoretical foundation of machine learning. Topics include the nonparametric estimation of the cumulative distribution function (Glivenko-Cantelli, Kolmogorov-Smirnov and Dvoretzky-Kiefer Wolfowitz theorems); nonparametric estimation of probability density functions (kernel methods); nonparametric regression; Bias-variance tradeoff and double descent phenomena; and VC-dimension and Rademacher complexity. Prerequisite(s): MAT 5002 .