ECE594(HDP) - High-dimensional probability

Instructor(s): Thrampoulidis


The goal of this course is to cover fundamental principles and techniques from high-dimensional probability and geometry. It is intended for graduate students in engineering, computer science, statistics and related areas, who are looking to acquire tools that have become essential to a variety of applications in data science and contemporary signal-processing. Topics to be covered include: concentration inequalities, sub-Gaussian processes, random vectors in high-dimensions, random matrices, comparison theorems. The tools will be used to study some stylized applications, such as sparse regression, matrix completion, dimension reduction, principal component analysis, and clustering.