Cancelled - Interaction screening and selection in Ultrahigh Dimensions

This talk has been cancelled. 

This talk is based on Dr Daoji Li's joint work with Yingying Fan, Yinfei Kong, Jinchi Lv and Zemin Zheng.

Understanding how features interact with each other is of paramount importance in many scientific discoveries and contemporary applications. Yet interaction identification becomes challenging even for a moderate number of covariates.

In this talk, Dr Li will first talk about recent advances in interaction screening and selection for ultrahigh dimensional data. Then he will introduce an efficient and flexible procedure for interaction identification for quadratic regression models in ultra-high dimensions.

The group's approach screens important interactions by examining only p features instead of all two-way interactions of order O(p^2). Under a fairly general framework, they establish that the proposed method enjoys sure screening property in screening and oracle inequalities in selection. They prove that their method admits an explicit bound on the false sign rate, which can be asymptotically vanishing. The method and theoretical results are supported by several numerical studies.

Speaker

Dr Daoji Li is an Assistant Professor in the Department of Information Systems and Decision Sciences at California State University Fullerton.