OSU Navigation Bar

The Ohio State University

Department of Statistics

Cockins Hall
rollover image OSU Statistics
            Home

design element

OSU Statistics

Home

News

Research & Consulting Groups

People

For Visitors

For Prospective Students

For Current Students & Faculty

Contact Us



rollover image

News

rollover image

Newsletter

rollover image

Seminars

Department of Statistics, The Ohio State University
Statistics and Biostatistics Colloquium Series

Hosted jointly with the Department of Computer Science and Engineering, The Ohio State University

2008 Chhotey Lal and Mohra Devi Rustagi Memorial Lecture

Fast Sparse Regression and Classification

Jerome H. Friedman
Stanford University

3:30PM - Thursday, May 29, 2008
Room 170, Eighteenth Avenue Bldg. (EA 170)

ABSTRACT

Regularized regression and classification methods fit a linear model to data, based on some loss criterion, subject to a constraint on the coefficient values. As special cases, ridge-regression, the lasso, and subset selection all use squared-error loss with different particular constraint choices. For large problems the general choice of loss/constraint combinations is usually limited by the computation required to obtain the corresponding solution estimates, especially when non convex constraints are used to induce very sparse solutions. A fast algorithm is presented that produces solutions that closely approximate those for any convex loss and a wide variety of convex and non convex constraints, permitting application to very large problems. The benefits of this generality are illustrated by examples.

Reception to follow the talk. Location to be announced.



If you have trouble accessing this page, or need an alternate format contact webmaster@stat.osu.edu.