A SUPERIOR TRAINING STRATEGY FOR THREE-LAYER FEEDFORWARD ARTIFICIAL NEURAL NETWORKS

Persistent Link:
http://hdl.handle.net/10150/614171
Title:
A SUPERIOR TRAINING STRATEGY FOR THREE-LAYER FEEDFORWARD ARTIFICIAL NEURAL NETWORKS
Author:
Hsu, Kuo-Lin; Gupta, Hoshin Vijai; Sorooshian, Soroosh
Affiliation:
Department of Hydrology & Water Resources, The University of Arizona
Publisher:
Department of Hydrology and Water Resources, University of Arizona (Tucson, AZ)
Issue Date:
1996-03
Rights:
Copyright © Arizona Board of Regents
Collection Information:
This title from the Hydrology & Water Resources Technical Reports collection is made available by the Department of Hydrology & Atmospheric Sciences and the University Libraries, University of Arizona. If you have questions about titles in this collection, please contact repository@u.library.arizona.edu.
Abstract:
A new algorithm is proposed for the identification of three-layer feedforward artificial neural networks. The algorithm, entitled LLSSIM, partitions the weight space into two major groups: the input- hidden and hidden -output weights. The input- hidden weights are trained using a multi -start SIMPLEX algorithm and the hidden -output weights are identified using a conditional linear- least- square estimation approach. Architectural design is accomplished by progressive addition of nodes to the hidden layer. The LLSSIM approach provides globally superior weight estimates with fewer function evaluations than the conventional back propagation (BPA) and adaptive back propagation (ABPA) strategies. Monte -carlo testing on the XOR problem, two function approximation problems, and a rainfall- runoff modeling problem show LLSSIM to be more effective, efficient and stable than BPA and ABPA.
Series/Report no.:
Technical Reports on Hydrology and Water Resources, No. 96-030
Sponsors:
This research was partially supported by grants from the Hydrologic Research Laboratory of the U.S. National Weather Service (Grant no. NA37WH0385), the NASA -EOS Interdisciplinary Research Program (IDP -88 -086), and the NOAA Research Program (NA16RC0119 -0). The first author greatly appreciates the fellowship support provided by the NASA Global Change Program (Grant No. NGT- 30045).

Full metadata record

DC FieldValue Language
dc.contributor.authorHsu, Kuo-Linen
dc.contributor.authorGupta, Hoshin Vijaien
dc.contributor.authorSorooshian, Sorooshen
dc.date.accessioned2016-06-22T18:38:51Z-
dc.date.available2016-06-22T18:38:51Z-
dc.date.issued1996-03-
dc.identifier.urihttp://hdl.handle.net/10150/614171-
dc.description.abstractA new algorithm is proposed for the identification of three-layer feedforward artificial neural networks. The algorithm, entitled LLSSIM, partitions the weight space into two major groups: the input- hidden and hidden -output weights. The input- hidden weights are trained using a multi -start SIMPLEX algorithm and the hidden -output weights are identified using a conditional linear- least- square estimation approach. Architectural design is accomplished by progressive addition of nodes to the hidden layer. The LLSSIM approach provides globally superior weight estimates with fewer function evaluations than the conventional back propagation (BPA) and adaptive back propagation (ABPA) strategies. Monte -carlo testing on the XOR problem, two function approximation problems, and a rainfall- runoff modeling problem show LLSSIM to be more effective, efficient and stable than BPA and ABPA.en
dc.description.sponsorshipThis research was partially supported by grants from the Hydrologic Research Laboratory of the U.S. National Weather Service (Grant no. NA37WH0385), the NASA -EOS Interdisciplinary Research Program (IDP -88 -086), and the NOAA Research Program (NA16RC0119 -0). The first author greatly appreciates the fellowship support provided by the NASA Global Change Program (Grant No. NGT- 30045).en
dc.language.isoen_USen
dc.publisherDepartment of Hydrology and Water Resources, University of Arizona (Tucson, AZ)en
dc.relation.ispartofseriesTechnical Reports on Hydrology and Water Resources, No. 96-030en
dc.rightsCopyright © Arizona Board of Regentsen
dc.sourceProvided by the Department of Hydrology and Water Resources.en
dc.titleA SUPERIOR TRAINING STRATEGY FOR THREE-LAYER FEEDFORWARD ARTIFICIAL NEURAL NETWORKSen_US
dc.typetexten
dc.typeTechnical Reporten
dc.contributor.departmentDepartment of Hydrology & Water Resources, The University of Arizonaen
dc.description.collectioninformationThis title from the Hydrology & Water Resources Technical Reports collection is made available by the Department of Hydrology & Atmospheric Sciences and the University Libraries, University of Arizona. If you have questions about titles in this collection, please contact repository@u.library.arizona.edu.en
All Items in UA Campus Repository are protected by copyright, with all rights reserved, unless otherwise indicated.