The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Abstract Details
|
Activity Number:
|
294
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 2, 2011 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
IMS
|
| Abstract - #301369 |
|
Title:
|
Coordinate Descent Algorithms for LASSO Problems: The Good, the Bad, and the Ugly
|
|
Author(s):
|
Noah Simon*+
|
|
Companies:
|
Stanford University
|
|
Address:
|
390 Serra Mall, Stanford University, Stanford, CA, 94305,
|
|
Keywords:
|
LASSO ;
coordinate descent ;
algorithms
|
|
Abstract:
|
Coordinate descent and block-wise descent algorithms, lauded for their simplicity and efficacy, have been used with much success to fit LASSO (and LASSO-like) problems. However there is some misunderstanding about when coordinate descent is appropriate. In this talk we examine aspects of lasso-like problems that make them particularly suited to coordinate descent and contrast these with more ill-suited problems. We look at potential difficulties with fitting by coordinate descent, and note when and how these difficulties can be overcome.
Parts of this talk will be based on joint work with Rob Tibshirani, and inspired by conversations with Jerome Friedman and Trevor Hastie.
|
The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.
Back to the full JSM 2011 program
|
2011 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.