Thông tin Talk 1

long_nguyen_2012.jpg

Speaker:  Nguyễn Xuân Long

University of Michigan

 

Talk title: Optimal transport and statistical inference for complex structured data.

Speaker: Nguyen Xuan Long, University of Michigan, Ann Arbor (báo cáo trực tiếp)

Time: 15:00-17:00, Tuesday, June 7, 2022.

Seminar: Hybrid seminar [Registration here]

Onsite: Speaker will give the talk at VIASM, 157 Chùa Láng, Đống Đa, Hà Nội. 

Online: Participation by the link:

Livestream: https://www.facebook.com/viasmeduvn

AbstractStatistical inference is concerned with understanding and making predictions about the observed data via the modeling of underlying probability distributions for the data population. Optimal transport theory is an optimization-based formalism for coupling one probability distribution with another. Optimal transport has a long and remarkable history, but it has also achieved significant new advances in the past several decades in various fields in mathematics.  In this talk I will describe two ways in which optimal transport is useful in statistics and machine learning: First, the modeling of distribution for functional data, data that can be modeled as realization of random functions. The second, perhaps deeper, role of optimal transport is in underzstanding and improving convergence behavior of numerous latent variable models in statistics such as mixture and hierarchical models. I'll describe some recent results and outline several open questions.

Bio: Nguyen Xuan Long is Professor of Statistics and of Electrical Engineering and Computer Science at the University of Michigan. His research interests lie in Bayesian nonparametric statistics, machine learning and optimal transport, and analysis of complex structured data. He has served as associate editor in several major journals, including the Annals of Statistics, Bayesian Analysis, SIAM Journal on Mathematics of Data Science and Journal of Machine Learning Research. He is a distinguished associate member of VIASM, and a fellow of the IMS (Institute of Mathematical Statistics).