Sijian Wang, Colloquium Speaker

Assistant Professor, Department of Biostatistics & Medical Informatics, University of Wisconsin at Madison
Date: 
Thursday, October 25, 2012 - 12:00am
Colloquium Title: 
Group and within group variable selection via convex penalty
Location: 
Reception at 3 p.m. in 241 B Schaeffer Hall / Talk at 3:30 p.m. in 140 Schaeffer Hall.

Abstract: In many scientific and engineering applications, predictors are naturally grouped, for example, in biological applications where assayed genes or proteins can be grouped by biological roles or biological pathways. When the group structures are available among predictors, people are usually interested in identifying both important groups and important variables within the selected groups. Among existing successful group variable selection methods, some methods fail to conduct the within group selection. Some methods are able to conduct both group and within group selection, but the corresponding objective function is non-convex, which may require extra numerical effort. In this talk, we will present a convex penalty for both group and within group variable selection.  We develop an efficient group-level coordinate descent algorithm for solving the corresponding optimization problem.  We also study the non-asymptotic properties of the estimates in the high-dimensional setting, where the number of predictors can be much larger than the sample size. Numerical results indicate that the proposed method works well in terms of both variable selection and prediction accuracy. We also apply the proposed method to American Cancer Society Breast Cancer Survivor dataset.