Date/Time Date(s) - 12/03/20243:30 am - 4:30 am
Location: KTH B105
Date/Time: Tuesday, March 12, 2024, 3.30 – 4.30 p.m. (will bring refreshments to KTH B105)
Speaker: Yanglei Song (Queen’s University)
Title: Concavity Test of Nonparametric Multiple Regression
Abstract: Principled nonparametric tests for regression curvature in R^d are often statistically and computationally challenging. We introduce stratified incomplete local simplex (SILS) tests for joint concavity of nonparametric multiple regression. We show that the SILS tests with suitable bootstrap calibration achieve simultaneous guarantees on dimension-free computational complexity, polynomial decay of the uniform error-in-size, and power consistency for general (global and local) alternatives. To establish these results, we develop a general theory for incomplete U-processes with stratified random sparse weights. Novel technical ingredients include maximal inequalities for the supremum of multiple incomplete U-processes. This is a joint work with Xiaohui Chen (USC) and Kengo Kato (Cornell).
Bio: Yanglei Song is an Assistant Professor in the Department of Mathematics and Statistics at Queen’s University. He obtained his PhD from the University of Illinois Urbana-Champaign in 2019 and joined Queen’s in the same year. His research interests include sequential decision-making problems, such as hypothesis testing, change detection, multi-armed bandits, and mathematical statistics, such as U-statistics and cut-point analysis.