Orthogonal least squares regression with tunable kernels

S. Chen*, X. X. Wang, D. J. Brown

*Corresponding author for this work

Research output: Contribution to journalArticle

Abstract

A novel technique is proposed to construct sparse regression models based on the orthogonal least squares method with tunable kernels. The proposed technique tunes the centre vector and diagonal covariance matrix of individual regressors by incrementally minimising the training mean square error using a guided random search algorithm, and it offers a state-of-the-art method for constructing very sparse models that generalise well.

Original languageEnglish
Pages (from-to)484-486
Number of pages3
JournalElectronics letters
Volume41
Issue number8
DOIs
Publication statusPublished - 14 Apr 2005

Fingerprint

Covariance matrix
Mean square error

Bibliographical note

© 2005 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Cite this

Chen, S. ; Wang, X. X. ; Brown, D. J. / Orthogonal least squares regression with tunable kernels. In: Electronics letters. 2005 ; Vol. 41, No. 8. pp. 484-486.
@article{7d132824a4414a1b88ccdf35c0a0b143,
title = "Orthogonal least squares regression with tunable kernels",
abstract = "A novel technique is proposed to construct sparse regression models based on the orthogonal least squares method with tunable kernels. The proposed technique tunes the centre vector and diagonal covariance matrix of individual regressors by incrementally minimising the training mean square error using a guided random search algorithm, and it offers a state-of-the-art method for constructing very sparse models that generalise well.",
author = "S. Chen and Wang, {X. X.} and Brown, {D. J.}",
note = "{\circledC} 2005 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.",
year = "2005",
month = "4",
day = "14",
doi = "10.1049/el:20050265",
language = "English",
volume = "41",
pages = "484--486",
journal = "Electronics letters",
issn = "0013-5194",
publisher = "IET",
number = "8",

}

Orthogonal least squares regression with tunable kernels. / Chen, S.; Wang, X. X.; Brown, D. J.

In: Electronics letters, Vol. 41, No. 8, 14.04.2005, p. 484-486.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Orthogonal least squares regression with tunable kernels

AU - Chen, S.

AU - Wang, X. X.

AU - Brown, D. J.

N1 - © 2005 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

PY - 2005/4/14

Y1 - 2005/4/14

N2 - A novel technique is proposed to construct sparse regression models based on the orthogonal least squares method with tunable kernels. The proposed technique tunes the centre vector and diagonal covariance matrix of individual regressors by incrementally minimising the training mean square error using a guided random search algorithm, and it offers a state-of-the-art method for constructing very sparse models that generalise well.

AB - A novel technique is proposed to construct sparse regression models based on the orthogonal least squares method with tunable kernels. The proposed technique tunes the centre vector and diagonal covariance matrix of individual regressors by incrementally minimising the training mean square error using a guided random search algorithm, and it offers a state-of-the-art method for constructing very sparse models that generalise well.

UR - http://www.scopus.com/inward/record.url?scp=18144409489&partnerID=8YFLogxK

U2 - 10.1049/el:20050265

DO - 10.1049/el:20050265

M3 - Article

AN - SCOPUS:18144409489

VL - 41

SP - 484

EP - 486

JO - Electronics letters

JF - Electronics letters

SN - 0013-5194

IS - 8

ER -