Bregman distance to L1 regularized logistic regression

Mithun Das Gupta; Thomas S. Huang

doi:10.1109/icpr.2008.4761922

Bregman distance to L₁ regularized logistic regression

Mithun Das Gupta, Thomas S. Huang

Coordinated Science Lab

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this work we investigate the relationship between Bregman distances and regularized Logistic Regression model. We convert L₁-regularized logistic regression (LR) into more general Bregman divergence framework and propose a primal-dual method based algorithm for learning the parameters of the model. The proposed method utilizes L₁ regularization to incorporate parameter sparsity into the divergence minimization scheme. We perform tests on public domain data sets and produce results which are amongst the best reported.

Original language	English (US)
Title of host publication	2008 19th International Conference on Pattern Recognition, ICPR 2008
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Print)	9781424421756
DOIs	https://doi.org/10.1109/icpr.2008.4761922
State	Published - 2008

Publication series

Name	Proceedings - International Conference on Pattern Recognition
ISSN (Print)	1051-4651

ASJC Scopus subject areas

Computer Vision and Pattern Recognition

Online availability

10.1109/icpr.2008.4761922

Library availability

Discover UIUC Full Text

Cite this

Bregman distance to L₁ regularized logistic regression. / Gupta, Mithun Das; Huang, Thomas S.
2008 19th International Conference on Pattern Recognition, ICPR 2008. Institute of Electrical and Electronics Engineers Inc., 2008. 4761922 (Proceedings - International Conference on Pattern Recognition).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

@inproceedings{886ee82492da4fbaaa9456b71cae6648,

title = "Bregman distance to L1 regularized logistic regression",

abstract = "In this work we investigate the relationship between Bregman distances and regularized Logistic Regression model. We convert L1-regularized logistic regression (LR) into more general Bregman divergence framework and propose a primal-dual method based algorithm for learning the parameters of the model. The proposed method utilizes L1 regularization to incorporate parameter sparsity into the divergence minimization scheme. We perform tests on public domain data sets and produce results which are amongst the best reported.",

author = "Gupta, {Mithun Das} and Huang, {Thomas S.}",

year = "2008",

doi = "10.1109/icpr.2008.4761922",

language = "English (US)",

isbn = "9781424421756",

series = "Proceedings - International Conference on Pattern Recognition",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2008 19th International Conference on Pattern Recognition, ICPR 2008",

address = "United States",

}

TY - GEN

T1 - Bregman distance to L1 regularized logistic regression

AU - Gupta, Mithun Das

AU - Huang, Thomas S.

PY - 2008

Y1 - 2008

N2 - In this work we investigate the relationship between Bregman distances and regularized Logistic Regression model. We convert L1-regularized logistic regression (LR) into more general Bregman divergence framework and propose a primal-dual method based algorithm for learning the parameters of the model. The proposed method utilizes L1 regularization to incorporate parameter sparsity into the divergence minimization scheme. We perform tests on public domain data sets and produce results which are amongst the best reported.

AB - In this work we investigate the relationship between Bregman distances and regularized Logistic Regression model. We convert L1-regularized logistic regression (LR) into more general Bregman divergence framework and propose a primal-dual method based algorithm for learning the parameters of the model. The proposed method utilizes L1 regularization to incorporate parameter sparsity into the divergence minimization scheme. We perform tests on public domain data sets and produce results which are amongst the best reported.

UR - http://www.scopus.com/inward/record.url?scp=77957939809&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77957939809&partnerID=8YFLogxK

U2 - 10.1109/icpr.2008.4761922

DO - 10.1109/icpr.2008.4761922

M3 - Conference contribution

AN - SCOPUS:77957939809

SN - 9781424421756

T3 - Proceedings - International Conference on Pattern Recognition

BT - 2008 19th International Conference on Pattern Recognition, ICPR 2008

PB - Institute of Electrical and Electronics Engineers Inc.

ER -