Kernel-based data modelling using orthogonal least squares selection with local regularisation
Kernel-based data modelling using orthogonal least squares selection with local regularisation
Combining orthogonal least squares (OLS) model selection with local regularisation or smoothing leads to efficient sparse kernel-based data modelling. By assigning each orthogonal weight in the regression model with an individual regularisation parameter, the ability for the OLS model selection to produce a very parsimonious model with excellent generalisation performance is greatly enhanced.
0 9533890 4 9
27-30
Chen, S.
9310a111-f79a-48b8-98c7-383ca93cbb80
September 2001
Chen, S.
9310a111-f79a-48b8-98c7-383ca93cbb80
Chen, S.
(2001)
Kernel-based data modelling using orthogonal least squares selection with local regularisation.
7th Annual Chinese Automation and Computer Science Conference in UK.
.
Record type:
Conference or Workshop Item
(Other)
Abstract
Combining orthogonal least squares (OLS) model selection with local regularisation or smoothing leads to efficient sparse kernel-based data modelling. By assigning each orthogonal weight in the regression model with an individual regularisation parameter, the ability for the OLS model selection to produce a very parsimonious model with excellent generalisation performance is greatly enhanced.
Other
cacauk01P.ps
- Other
More information
Published date: September 2001
Additional Information:
Presented at 7th Annual Chinese Automation and Computer Science Conference in UK (Nottingham) , Sept.22, 2001
Venue - Dates:
7th Annual Chinese Automation and Computer Science Conference in UK, 2001-09-01
Organisations:
Southampton Wireless Group
Identifiers
Local EPrints ID: 255977
URI: http://eprints.soton.ac.uk/id/eprint/255977
ISBN: 0 9533890 4 9
PURE UUID: e3d0af4b-9daa-4086-aa25-2dc664d4f9b0
Catalogue record
Date deposited: 24 Sep 2001
Last modified: 14 Mar 2024 05:37
Export record
Contributors
Author:
S. Chen
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics