Wang, Xiangyu, Dusnon, David and Leng, Chenlei (2016) No penalty no tears : least squares in high-dimensional linear models. In: 33rd International Conference on Machine Learning, New York City, USA, 19-24 Jun 2016. Published in: Proceedings of the 33rd International Conference on Machine Learning (ICML 2016) pp. 1814-1822. ISBN 1938-7228.

PDF WRAP_1273119-st-170516-1506.02222v1.pdf - Accepted Version - Requires a PDF viewer. Download (3670Kb)

Request Changes to record.

Abstract

Ordinary least squares (OLS) is the default method for fitting linear models, but is not applicable for problems with dimensionality larger than the sample size. For these problems, we advocate the use of a generalized version of OLS motivated by ridge regression, and propose two novel three-step algorithms involving least squares fitting and hard thresholding. The algorithms are methodologically simple to understand intuitively, computationally easy to implement efficiently, and theoretically appealing for choosing models consistently. Numerical exercises comparing our methods with penalization-based approaches in simulations and data analyses illustrate the great potential of the proposed algorithms.

Item Type:	Conference Item (Paper)
Divisions:	Faculty of Science > Statistics
Journal or Publication Title:	Proceedings of the 33rd International Conference on Machine Learning (ICML 2016)
ISBN:	1938-7228
Official Date:	2016
Dates:	Date Event 2016 Published 24 April 2016 Accepted
Date of first compliant deposit:	20 May 2016
Page Range:	pp. 1814-1822
Status:	Peer Reviewed
Publication Status:	Published
Conference Paper Type:	Paper
Title of Event:	33rd International Conference on Machine Learning
Type of Event:	Conference
Location of Event:	New York City, USA
Date(s) of Event:	19-24 Jun 2016
Related URLs:	Publisher
Open Access Version:	Publisher

Request changes or add full text files to a record

Repository staff actions (login required)

View Item

Downloads