Orthogonal Forward Regression based on Directly Maximizing Model Generalization Capability
Orthogonal Forward Regression based on Directly Maximizing Model Generalization Capability
The paper introduces a construction algorithm for sparse kernel modelling using the leave-one-out test score also known as the PRESS (Predicted REsidual Sums of Squares) statistic. An efficient subset model selection procedure is developed in the orthogonal forward regression framework by incrementally maximizing the model generalization capability to construct sparse models with good generalization properties. The proposed algorithm achieves a fully automated model construction without resort to any other validation data set for costly model evaluation.
0 9533890 6 5
251-256
Chen, S.
9310a111-f79a-48b8-98c7-383ca93cbb80
Hong, X.
b8f251c3-e142-4555-a54c-c504de966b03
2003
Chen, S.
9310a111-f79a-48b8-98c7-383ca93cbb80
Hong, X.
b8f251c3-e142-4555-a54c-c504de966b03
Chen, S. and Hong, X.
(2003)
Orthogonal Forward Regression based on Directly Maximizing Model Generalization Capability.
9th Annual Conference of Chinese Automation and Computing Society in UK, University of Luton, Luton, United Kingdom.
.
Record type:
Conference or Workshop Item
(Other)
Abstract
The paper introduces a construction algorithm for sparse kernel modelling using the leave-one-out test score also known as the PRESS (Predicted REsidual Sums of Squares) statistic. An efficient subset model selection procedure is developed in the orthogonal forward regression framework by incrementally maximizing the model generalization capability to construct sparse models with good generalization properties. The proposed algorithm achieves a fully automated model construction without resort to any other validation data set for costly model evaluation.
Other
cacsuk03a.ps
- Other
Other
cacsuk03aP.ps
- Other
Text
cacsuk03a.pdf
- Other
Text
cacsuk03aP.pdf
- Other
More information
Published date: 2003
Additional Information:
Event Dates: 20th September 2003
Venue - Dates:
9th Annual Conference of Chinese Automation and Computing Society in UK, University of Luton, Luton, United Kingdom, 2003-09-20
Organisations:
Southampton Wireless Group
Identifiers
Local EPrints ID: 259254
URI: http://eprints.soton.ac.uk/id/eprint/259254
ISBN: 0 9533890 6 5
PURE UUID: 97466057-1860-4eb7-8fa0-b17ca01702bc
Catalogue record
Date deposited: 14 Apr 2004
Last modified: 14 Mar 2024 06:21
Export record
Contributors
Author:
S. Chen
Author:
X. Hong
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