line graphs displayed across a digital screen


Dr. Xianyang Zhang

Dr. Xianyang Zhang, Associate Professor, Department of Statistics, Texas A&M University Presents Kernel Two-Sample Metrics in High-Dimension

Date: Friday, April 15, 2022

Motivated by the increasing use of kernel-based metrics for high dimensional and large-scale data, we study the asymptotic behavior of kernel two-sample tests when the dimension and sample sizes both diverge to infinity. We focus on the maximum mean discrepancy (MMD), including MMD with the Gaussian kernel and the Laplace kernel, and the energy distance as special cases. We derive asymptotic expansions of the kernel two-sample statistics, based on which we establish the central limit theorem (CL

Dr. Michael Kosorok

Dr. Michael Kosorok, W.R. Kenan Jr. Distinguished Professor, Department of Biostatistics, Gillings School of Global Public Health, Department of Statistics and Operations Research, University of North Carolina at Chapel Hill presents Nonparametric Reinforcement Learning for Survival Outcomes

Date: Friday, March 25, 2022

In some disease settings, such as cancer, there are several stages in the course of treatment where decisions on what treatment to select are made. A key goal of precision medicine in these settings is to determine optimal treatment based on patient status and history. In this presentation, we discuss a flexible new approach to making these determinations based on observed data, with the goal of maximizing a right-censored event time such as overall survival. Our method combines nonparametric ra

Dr. Yinchu Zhu

Dr. Yinchu Zhu, Assistant Professor of Economics at Brandeis University presents How well can we learn large factor models without assuming strong factors?

Date: Friday, March 4, 2022

In this paper, we consider the problem of learning models with a latent factor structure. The focus is to find what is possible and what is impossible if the usual strong factor condition is not imposed. We study the minimax rate and adaptivity issues in two problems: pure factor models and panel regression with interactive fixed effects. For pure factor models, if the number of factors is known, we develop adaptive estimation and inference procedures that attain the minimax rate. However, when

Dr. David Kepplinger

Dr. David Kepplinger, George Mason University presents Reliable Feature Selection and Prediction under Adverse Contamination

Date: Friday, December 3, 2021

Biomarker discovery studies have seen a proliferation with increasingly affordable high-throughput proteomics and genome sequencing. Given the abundance of such studies, the number of generalizable and clinically relevant discoveries is lacking. In particular, many discovery studies are based on samples of limited size and from potentially heterogenous populations, while at the same time hundreds or even thousands of genes are sequenced. In addition to the heterogeneity of the population, this i

Dr. Jordan Rodu

Dr. Jordan Rodu, University of Virginia presents Understanding Outcome Reasoning and the Rise of Machine Learning

Date: Friday, November 12, 2021

A new form of reasoning about data has emerged over the past several decades that is almost entirely responsible for the dramatic gains seen in the machine learning and AI communities. The implicit assumptions of this secret ingredient have important implications for its suitability for use in various types of problems, but these assumptions have gone all but unacknowledged in the ML and statistics communities. In this talk, we provide language to talk about this new form of reasoning—which we c

Dr. Yuan Liao

Dr. Yuan Liao, Rutgers University presents Structural Deep Learning in a Conditional Asset Pricing

Date: Friday, November 5, 2021

We develop new nonparametric methodology for estimating conditional asset pricing models using deep neural networks, by employing time-varying conditional information on alphas and betas carried by firm-specific characteristics. Contrary to many applications of neural networks in economics, we can open the “black box” of machine learning predictions, and provide an economic interpretation of the successful predictions obtained from neural networks. We formally establish the asymptotic theory of

Check out our past events.