Please use this identifier to cite or link to this item: https://doi.org/10.1109/TNN.2004.824266
DC FieldValue
dc.titleAn efficient method for computing leave-one-out error in support vector machines with Gaussian kernels
dc.contributor.authorLee, M.M.S.
dc.contributor.authorKeerthi, S.S.
dc.contributor.authorOng, C.J.
dc.contributor.authorDeCoste, D.
dc.date.accessioned2014-06-17T06:11:29Z
dc.date.available2014-06-17T06:11:29Z
dc.date.issued2004-05
dc.identifier.citationLee, M.M.S., Keerthi, S.S., Ong, C.J., DeCoste, D. (2004-05). An efficient method for computing leave-one-out error in support vector machines with Gaussian kernels. IEEE Transactions on Neural Networks 15 (3) : 750-757. ScholarBank@NUS Repository. https://doi.org/10.1109/TNN.2004.824266
dc.identifier.issn10459227
dc.identifier.urihttp://scholarbank.nus.edu.sg/handle/10635/59442
dc.description.abstractIn this paper, we give an efficient method for computing the leave-one-out (LOO) error for support vector machines (SVMs) with Gaussian kernels quite accurately. It is particularly suitable for iterative decomposition methods of solving SVMs. The importance of various steps of the method is illustrated in detail by showing the performance on six benchmark datasets. The new method often leads to speedups of 10-50 times compared to standard LOO error computation. It has good promise for use in hyperparameter tuning and model comparison.
dc.description.urihttp://libproxy1.nus.edu.sg/login?url=http://dx.doi.org/10.1109/TNN.2004.824266
dc.sourceScopus
dc.subjectLeave-one-out (LOO) error
dc.subjectSupport vector machines (SVMs)
dc.typeArticle
dc.contributor.departmentMECHANICAL ENGINEERING
dc.description.doi10.1109/TNN.2004.824266
dc.description.sourcetitleIEEE Transactions on Neural Networks
dc.description.volume15
dc.description.issue3
dc.description.page750-757
dc.description.codenITNNE
dc.identifier.isiut000221483700018
Appears in Collections:Staff Publications

Show simple item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

59
checked on Aug 17, 2019

WEB OF SCIENCETM
Citations

41
checked on Aug 9, 2019

Page view(s)

58
checked on Aug 18, 2019

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.