Gaussian kernel approximation algorithm for feedforward neural network design

A Gaussian kernel approximation algorithm for a feedforward neural network is presented. The approach used by the algorithm, which is based on a constructive learning algorithm, is to create the hidden units directly so that automatic design of the architecture of neural networks can be carried out....

Full description

Saved in:
Bibliographic Details
Main Authors: Ananta Srisuphab, Jarernsri L. Mitrpanont
Other Authors: Mahidol University
Format: Article
Published: 2018
Subjects:
Online Access:https://repository.li.mahidol.ac.th/handle/123456789/27767
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Mahidol University
id th-mahidol.27767
record_format dspace
spelling th-mahidol.277672018-09-13T13:47:34Z Gaussian kernel approximation algorithm for feedforward neural network design Ananta Srisuphab Jarernsri L. Mitrpanont Mahidol University Mathematics A Gaussian kernel approximation algorithm for a feedforward neural network is presented. The approach used by the algorithm, which is based on a constructive learning algorithm, is to create the hidden units directly so that automatic design of the architecture of neural networks can be carried out. The algorithm is defined using the linear summation of input patterns and their randomized input weights. Hidden-layer nodes are defined so as to partition the input space into homogeneous regions, where each region contains patterns belonging to the same class. The largest region is used to define the center of the corresponding Gaussian hidden nodes. The algorithm is tested on three benchmark data sets of different dimensionality and sample sizes to compare the approach presented here with other algorithms. Real medical diagnoses and a biological classification of mushrooms are used to illustrate the performance of the algorithm. These results confirm the effectiveness of the proposed algorithm. © 2009 Elsevier Inc. All rights reserved. 2018-09-13T06:47:34Z 2018-09-13T06:47:34Z 2009-12-01 Article Applied Mathematics and Computation. Vol.215, No.7 (2009), 2686-2693 10.1016/j.amc.2009.09.008 00963003 2-s2.0-70350721632 https://repository.li.mahidol.ac.th/handle/123456789/27767 Mahidol University SCOPUS https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=70350721632&origin=inward
institution Mahidol University
building Mahidol University Library
continent Asia
country Thailand
Thailand
content_provider Mahidol University Library
collection Mahidol University Institutional Repository
topic Mathematics
spellingShingle Mathematics
Ananta Srisuphab
Jarernsri L. Mitrpanont
Gaussian kernel approximation algorithm for feedforward neural network design
description A Gaussian kernel approximation algorithm for a feedforward neural network is presented. The approach used by the algorithm, which is based on a constructive learning algorithm, is to create the hidden units directly so that automatic design of the architecture of neural networks can be carried out. The algorithm is defined using the linear summation of input patterns and their randomized input weights. Hidden-layer nodes are defined so as to partition the input space into homogeneous regions, where each region contains patterns belonging to the same class. The largest region is used to define the center of the corresponding Gaussian hidden nodes. The algorithm is tested on three benchmark data sets of different dimensionality and sample sizes to compare the approach presented here with other algorithms. Real medical diagnoses and a biological classification of mushrooms are used to illustrate the performance of the algorithm. These results confirm the effectiveness of the proposed algorithm. © 2009 Elsevier Inc. All rights reserved.
author2 Mahidol University
author_facet Mahidol University
Ananta Srisuphab
Jarernsri L. Mitrpanont
format Article
author Ananta Srisuphab
Jarernsri L. Mitrpanont
author_sort Ananta Srisuphab
title Gaussian kernel approximation algorithm for feedforward neural network design
title_short Gaussian kernel approximation algorithm for feedforward neural network design
title_full Gaussian kernel approximation algorithm for feedforward neural network design
title_fullStr Gaussian kernel approximation algorithm for feedforward neural network design
title_full_unstemmed Gaussian kernel approximation algorithm for feedforward neural network design
title_sort gaussian kernel approximation algorithm for feedforward neural network design
publishDate 2018
url https://repository.li.mahidol.ac.th/handle/123456789/27767
_version_ 1763488868855709696