Please use this identifier to cite or link to this item: https://ktisis.cut.ac.cy/handle/10488/182
DC FieldValueLanguage
dc.contributor.authorTsapatsoulis, Nicolasen
dc.contributor.authorWallace, Manolisen
dc.contributor.authorKasderidis, Stathisen
dc.contributor.otherΤσαπατσούλης, Νικόλας-
dc.date.accessioned2009-05-27T10:50:28Zen
dc.date.accessioned2013-05-16T14:08:45Z-
dc.date.accessioned2015-12-02T12:30:06Z-
dc.date.available2009-05-27T10:50:28Zen
dc.date.available2013-05-16T14:08:45Z-
dc.date.available2015-12-02T12:30:06Z-
dc.date.issued2003en
dc.identifier.citationArtificial Neural Networks and Neural Information Processing — ICANN/ICONIP 2003, p.173en
dc.identifier.isbn9783540404088en
dc.identifier.urihttp://ktisis.cut.ac.cy/handle/10488/182en
dc.descriptionICANN/ICONIP 2003,(2003,Istanbul,Turkey)en
dc.description.abstractAdaptivity to non-stationary contexts is a very important property for intelligent systems in general, as well as to a variety of applications of knowledge based systems in era of “ambient intelligence”. In this paper we present a modified Resource Allocating Network architecture that allows for online adaptation and knowledge modelling through its adaptive structure. As in any neural network system proper parameter initialization reduces training time and effort. However, in RAN architectures, proper parameter initialization also leads to compact modelling (less hidden nodes) of the process under examination, and consequently to better generalization. In the cases of high-dimensional data parameter initialization is both difficult and time consuming. In the proposed scheme a high – dimensional, unsupervised clustering method is used to properly initialize the RAN architecture. Clusters correspond to the initial nodes of RAN, while output layer weights are also extracted from the clustering procedure. The efficiency of the proposed method has been tested on several classes of publicly available data (iris, ionosphere, etc.)en
dc.formatpdfen
dc.language.isoenen
dc.publisherSpringer Berlin / Heidelbergen
dc.relation.ispartofseriesArtificial Neural Networks and Neural Information Processing — ICANN/ICONIP 2003;en
dc.rights© Springeren
dc.subjectNeural networks (Computer science)--Congressesen
dc.titleImproving the Performance of Resource Allocation Networks through Hierarchical Clustering of High-Dimensional Dataen
dc.typeBook Chapteren
dc.collaborationNational Technical University Of Athens-
dc.collaborationKing's College London-
dc.countryGreece-
dc.countryUnited Kingdom-
dc.identifier.doi10.1007/3-540-44989-2_22en
dc.dept.handle123456789/54en
item.grantfulltextopen-
item.fulltextWith Fulltext-
item.languageiso639-1other-
crisitem.author.deptDepartment of Communication and Internet Studies-
crisitem.author.facultyFaculty of Communication and Media Studies-
crisitem.author.orcid0000-0002-6739-8602-
crisitem.author.parentorgFaculty of Communication and Media Studies-
Appears in Collections:Κεφάλαια βιβλίων/Book chapters
Files in This Item:
File Description SizeFormat
Tsapatsoulis_Improving the Performance of Resource.pdf115.04 kBAdobe PDFView/Open
Show simple item record

Page view(s)

80
Last Week
2
Last month
4
checked on Sep 19, 2019

Download(s)

44
checked on Sep 19, 2019

Google ScholarTM

Check

Altmetric


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