Advanced

Elastic Net Regularized Logistic Regression using Cubic Majorization

Nilsson, Mikael LU (2014) 22nd International Conference on Pattern Recognition (ICPR 2014) In 2014 22nd International Conference on Pattern Recognition (ICPR) p.3446-3451
Abstract
In this work, a coordinate solver for elastic net regularized logistic regression is proposed. In particular, a method based on majorization maximization using a cubic function is derived. This to reliably and accurately optimize the objective function at each step without resorting to line search. Experiments show that the proposed solver is comparable to, or improves, state-of-the-art solvers. The proposed method is simpler, in the sense that there is no need for any line search, and can directly be used for small to large scale learning problems with elastic net regularization.
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
in
2014 22nd International Conference on Pattern Recognition (ICPR)
pages
3446 - 3451
publisher
IEEE--Institute of Electrical and Electronics Engineers Inc.
conference name
22nd International Conference on Pattern Recognition (ICPR 2014)
external identifiers
  • wos:000359818003097
  • scopus:84919934120
ISSN
1051-4651
DOI
10.1109/ICPR.2014.593
language
English
LU publication?
yes
id
a4b2e4da-8bef-4cbd-bdbd-3fec275af927 (old id 7972402)
date added to LUP
2015-09-23 13:25:57
date last changed
2017-05-28 04:13:45
@inproceedings{a4b2e4da-8bef-4cbd-bdbd-3fec275af927,
  abstract     = {In this work, a coordinate solver for elastic net regularized logistic regression is proposed. In particular, a method based on majorization maximization using a cubic function is derived. This to reliably and accurately optimize the objective function at each step without resorting to line search. Experiments show that the proposed solver is comparable to, or improves, state-of-the-art solvers. The proposed method is simpler, in the sense that there is no need for any line search, and can directly be used for small to large scale learning problems with elastic net regularization.},
  author       = {Nilsson, Mikael},
  booktitle    = {2014 22nd International Conference on Pattern Recognition (ICPR)},
  issn         = {1051-4651},
  language     = {eng},
  pages        = {3446--3451},
  publisher    = {IEEE--Institute of Electrical and Electronics Engineers Inc.},
  title        = {Elastic Net Regularized Logistic Regression using Cubic Majorization},
  url          = {http://dx.doi.org/10.1109/ICPR.2014.593},
  year         = {2014},
}