When
Thursday, November 19, 2026, at 2:00 p.m.
Helen Zhang
Chair, Statistics and Data Science Interdisciplinary Program
Professor, Mathematics
University of Arizona
"Dynamic Supervised PCA for High-Dimensional Classification with Evolving Distributions"
Engineering, Room 301
Abstract: High-dimensional classification often suffers when class distributions shift over time, as static subspace projections and decision boundaries become ineffective. We propose a framework for dynamic high-dimensional classification that uses supervised PCA via kernel smoothing to reduce dimensionality and construct adaptive decision boundaries that evolve with changing distributions. We develop supporting theory and algorithms for both linear and quadratic discriminant analysis and demonstrate the approach’s effectiveness through simulations and real data applications.
Bio: Hao (Helen) Zhang is a professor of mathematics and chair of the Statistics and Data Science Interdisciplinary Program at the University of Arizona. Her research focuses on nonparametric modeling, statistical machine learning and high-dimensional data analysis, with applications in biomedical and engineering sciences. She has published more than 100 research articles and received extensive funding from NSF, the NIH and the NSA, including the NSF CAREER and NSF TRIPODS awards. Zhang has served as editor-in-chief of STAT and as associate editor of JASA and JRSSB. She is a fellow of the IMS and the ASA and an IMS Medallion lecturer.