Tools from modern high-dimensional probability and statistics, with applications to data science, machine learning, and algorithms. Special attention will be given to problems that arise from the analysis of matrix, graph and tensor data.

Mathematical tools:

Concentration inequalities

Random matrix theory

Gaussian comparison

Algorithmic tools:

Spectral methods

SDP relaxations

Message passing

Problems

Clustering;

Matrix completion

Graph localization

Dimensionality reduction and manifold learning

Class Times and Locations

Mon-Wed, 4:00-5:20pm

First lecture on Monday, January 11