The focus of this course is on foundations of (machine) learning theory based on statistical concepts. Learning Theory is concerned with extracting information from the data with or without supervision. Key tasks include prediction, classification, clustering, (low dimensional) structure identification, and related statistical inferential issues, with probabilistic guarantee of performance. This course is designed ideally for graduate students with background in Statistics core courses. The prerequisites are not hard set, but the student needs to be able to understand and apply basic results from real analysis, (weak) convergence, measure/probability theory, and statistical inference. Topics include: parametric and nonparametric function estimation methodology, resampling methods and uncertainty quantification, statistical methods for dimension reduction and structure identification, classification, clustering and prediction theory.
Section 01Topics in Statistics
INSTRUCTOR: LahiriView Course Listing