Talk Title

Statistical Inference with Stochastic Gradient Algorithms

Abstract

This talk will focus on the asymptotic properties of stochastic gradient methods used as sampling algorithms. We present a Bernstein--von Mises-like theorem for the scaling limit of the paths of stochastic gradient algorithms, showing they converge to an Ornstein-Uhlenbeck process. Then, using the large sample asymptotics, we demonstrate how to properly tune SGAs for various desiderata, including matching the asymptotics of the posterior distribution. 

Bio

Jeff is a postdoc at the University of Chicago's Data Science Institute. He is interested in a diverse set of research topics including statistical learning theory, computational statistics, and sequential decision making. Jeff completed his PhD in statistical sciences at the University of Toronto. Starting next academic year, Jeff will be joining the University of Waterloo's department of statistics and actuarial science as an assistant professor.