Statistics Seminar: Gourab Mukherjee

Speaker: Gourab Mukherjee

Title: Improved Nonparametric Empirical Bayes Estimation using Transfer Learning

Abstract: We consider the problem of estimating a multivariate normal mean in the presence of possibly useful auxiliary variables. The traditional nonparametric empirical Bayes (NEB) framework provides an elegant interface to pool information across dimensions and facilitates the construction of effective shrinkage estimators. Such estimators can be further improved by incorporating pertinent information from the auxiliary variables. However, detecting and assimilating possibly useful information from auxiliary variables to shrinkage estimators is difficult. Here, we develop a new methodology that can transfer useful information from multiple auxiliary variables and yield improved Tweedie-type NEB estimators. Our method uses convex optimization to directly estimate the gradient of the log-density through an embedding in the reproducing kernel Hilbert space induced by the Stein's discrepancy metric. We establish asymptotic optimality of the resultant estimator. We precisely tabulate the improvements in the estimation error as well as the deterioration in the learning rate as we inspect an increasing number of auxiliary variables. We demonstrate the competitive optimality of our method over existing NEB approaches through simulation experiments and in real data settings. This is joint work with Jiajun Luo and Wenguang Sun.

Bio: I am an Assistant Professor of Data Sciences and Operations in the University of Southern California's Marshall School of Business. I received my PhD in Statistics from Stanford University and had master's and undergraduate degrees in statistics from Indian Statistical Institute, Kolkata. My research interests are in mathematical statistics and in contemporary applications of statistics to virology and marketing. 

A virtual social will take place from 11:30-12:00, following this seminar. To join, visit z.umn.edu/STATSeminarSocial
NOTE: You will need to use a Chrome or Firefox browser to access this virtual platform. The link may not open if you sign in using a mobile device or tablet. The virtual space will not open until approximately 11:30.