File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Use of a quasi-Newton method in a feedforward neural network construction algorithm

TitleUse of a quasi-Newton method in a feedforward neural network construction algorithm
Authors
Issue Date1995
Citation
Ieee Transactions On Neural Networks, 1995, v. 6 n. 1, p. 273-277 How to Cite?
AbstractInterest in algorithms which dynamically construct neural networks has been growing in recent years. This paper describes an algorithm for constructing a single hidden layer feedforward neural network. A distinguishing feature of this algorithm is that it uses the quasi-Newton method to minimize the sequence of error functions associated with the growing network. Experimental results-indicate that the algorithm is very efficient and robust. The algorithm was tested on two test problems. The first was the n-bit parity problem and the second was the breast cancer diagnosis problem from the University of Wisconsin Hospitals. For the n-bit parity problem, the algorithm was able to construct neural network having less than n hidden units that solved the problem for n = 4, ···, 7. For the cancer diagnosis problem, the neural networks constructed by the algorithm had small number of hidden units and high accuracy rates on both the training data and the testing data.
Persistent Identifierhttp://hdl.handle.net/10722/152252
ISSN
2011 Impact Factor: 2.952
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorSetiono, Rudyen_US
dc.contributor.authorHui, Lucas Chi Kwongen_US
dc.date.accessioned2012-06-26T06:36:45Z-
dc.date.available2012-06-26T06:36:45Z-
dc.date.issued1995en_US
dc.identifier.citationIeee Transactions On Neural Networks, 1995, v. 6 n. 1, p. 273-277en_US
dc.identifier.issn1045-9227en_US
dc.identifier.urihttp://hdl.handle.net/10722/152252-
dc.description.abstractInterest in algorithms which dynamically construct neural networks has been growing in recent years. This paper describes an algorithm for constructing a single hidden layer feedforward neural network. A distinguishing feature of this algorithm is that it uses the quasi-Newton method to minimize the sequence of error functions associated with the growing network. Experimental results-indicate that the algorithm is very efficient and robust. The algorithm was tested on two test problems. The first was the n-bit parity problem and the second was the breast cancer diagnosis problem from the University of Wisconsin Hospitals. For the n-bit parity problem, the algorithm was able to construct neural network having less than n hidden units that solved the problem for n = 4, ···, 7. For the cancer diagnosis problem, the neural networks constructed by the algorithm had small number of hidden units and high accuracy rates on both the training data and the testing data.en_US
dc.languageengen_US
dc.relation.ispartofIEEE Transactions on Neural Networksen_US
dc.titleUse of a quasi-Newton method in a feedforward neural network construction algorithmen_US
dc.typeArticleen_US
dc.identifier.emailHui, Lucas Chi Kwong:hui@cs.hku.hken_US
dc.identifier.authorityHui, Lucas Chi Kwong=rp00120en_US
dc.description.naturelink_to_subscribed_fulltexten_US
dc.identifier.doi10.1109/72.363426en_US
dc.identifier.scopuseid_2-s2.0-0029185114en_US
dc.identifier.volume6en_US
dc.identifier.issue1en_US
dc.identifier.spage273en_US
dc.identifier.epage277en_US
dc.identifier.isiWOS:A1995QA72100030-
dc.publisher.placeUnited Statesen_US
dc.identifier.scopusauthoridSetiono, Rudy=7005033162en_US
dc.identifier.scopusauthoridHui, Lucas Chi Kwong=8905728300en_US

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats