Improving the performance of cascade correlation neural networks on multimodal functions

Simple item page
dc.contributor.authorRiley, Mike J. W.
dc.contributor.authorThompson, Christopher P.
dc.contributor.authorJenkins, Karl W.
dc.date.accessioned2019-11-11T16:02:41Z
dc.date.available2019-11-11T16:02:41Z
dc.date.issued2010-12-31
dc.description.abstractIntrinsic qualities of the cascade correlation algorithm make it a popular choice for many researchers wishing to utilize neural networks. Problems arise when the outputs required are highly multimodal over the input domain. The mean squared error of the approximation increases significantly as the number of modes increases. By applying ensembling and early stopping, we show that this error can be reduced by a factor of three. We also present a new technique based on subdivision that we call patchworking. When used in combination with early stopping and ensembling the mean improvement in error is over 10 in some cases.en_UK
dc.identifier.citationRiley MJW, Thompson CP, Jenkins KW. (2010) Improving the performance of cascade correlation neural networks on multimodal functions. In: 2010 World congress on engineering (WCE 2010), London, 30 June - 2 July 2010en_UK
dc.identifier.isbn978-988-18210-8-9
dc.identifier.urihttps://www.iaeng.org/publication/details/WCE2010_Proc_III.html
dc.identifier.urihttps://dspace.lib.cranfield.ac.uk/handle/1826/14708
dc.language.isoenen_UK
dc.publisherNewswood Limiteden_UK
dc.rightsAttribution-NonCommercial 4.0 International*
dc.rights.urihttp://creativecommons.org/licenses/by-nc/4.0/*
dc.titleImproving the performance of cascade correlation neural networks on multimodal functionsen_UK
dc.typeConference paperen_UK

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Improving_the_performance_of_cascade_correlation_neural-2010.pdf
Size:
999.21 KB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.63 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Staff publications (SATM)