Professor Olga Klopp, ESSEC Business School and CREST

"Adaptive density estimation under low-rank constraints"
Friday, 22 November 2024. 16:00-17:30
Online

Abstract

In this talk, we address the challenge of bivariate probability density estimation under low-rank constraints for both discrete and continuous distributions. For discrete distributions, we model the target as a low-rank probability matrix. In the continuous case, we assume the density function is Lipschitz continuous over an unknown compact rectangular support and can be decomposed into a sum of K separable components, each represented as a product of two one-dimensional functions. We introduce an estimator that leverages these low-rank constraints, achieving significantly improved convergence rates. We also derive lower bounds for both discrete and continuous cases, demonstrating that our estimators achieve minimax optimal convergence rates within logarithmic factors.

Bio

Olga Klopp is currently Professor of Statistic of the Department of Information Systems, Data Analytics and Operations at ESSEC Business School and Academic Director of the ESSEC- CentraleSupélec and ESSEC-ENSAE Double Degrees. She was nominated Beaufort Visiting Fellow at St John’s College at Cambridge University in 2023 and obtained Fulbright Scholar Award - National Program in 2024. Her field of research is in statistical learning and especially in auto-regressive models, missing data, and network analysis. She is a member of the ESSEC Risk Research Center CREAR and of CREST, EN- SAE, Institut Polytechnique de Paris. She has published her work, inter alia, in the Journal of the American Statistical Association, NeurIPS, Annals of Statistics and Bernoulli Journal. She has been a visiting scholar at, inter alia, Potsdam University, Berkeley University, Cambridge University, SAMSI Institute, Research Triangle Park USA .


For further information, please contact business-seminar-series@glasgow.ac.uk.

We foster a positive and productive environment for seminars through our Code of conduct.

First published: 14 November 2024