Active Learning for Probabilistic Neural Networks
Identifieur interne : 000A38 ( Main/Exploration ); précédent : 000A37; suivant : 000A39Active Learning for Probabilistic Neural Networks
Auteurs : Bülent Bolat [Turquie] ; Tülay Y Ld R M [Turquie]Source :
- Lecture Notes in Computer Science [ 0302-9743 ] ; 2005.
English descriptors
- Teeft :
- Active learner, Active sampling, Active selection, Algorithm, Basic architecture, Bayesian networks, Better training, Biggest class, Binary, Binary output value, Binary value, Bolat, Breast mass, Bupa, Continuous features, Data collection, Data exchange, Data exchange method, Data replication, Database, Dataset, Datasets, Decision boundary, Decision rule, Digitized image, Entire sample space, Exchange process, Generalization ability, Generalization error, Generalization performance, Glass dataset, Good selection, Hyper planes, Ieee trans, Input layer, Input space, Intelligent systems, International conference, Learner, Less instance numbers, Loss function, Loss functions, Maximum error, Multi layer perceptron, Network toolbox, Neural, Neural information processing systems, Neural network, Neural networks, Neuron, Original location, Other hand, Output layer, Parameter estimation, Parzen windows, Passive observer, Pattern layer neurons, Pattern vector, Probabilistic, Probability density functions, Proc, Redundant instances, Replication, Same training, Sample space, Square error, Summation layer, Support vector machines, Test accuracies, Test accuracy, Training data, Training data boosts, Training examples, Training instance, Training instances, Training pattern, Training patterns, Training sets, Useful training data selection criteria.
Abstract
Abstract: In many neural network applications, the selection of best training set to represent the entire sample space is one of the most important problems. Active learning algorithms in the literature for neural networks are not appropriate for Probabilistic Neural Networks (PNN). In this paper, a new active learning method is proposed for PNN. The method was applied to several benchmark problems.
Url:
DOI: 10.1007/11539087_13
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 001A61
- to stream Istex, to step Curation: 001937
- to stream Istex, to step Checkpoint: 000837
- to stream Main, to step Merge: 000A37
- to stream Main, to step Curation: 000A38
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Active Learning for Probabilistic Neural Networks</title>
<author><name sortKey="Bolat, Bulent" sort="Bolat, Bulent" uniqKey="Bolat B" first="Bülent" last="Bolat">Bülent Bolat</name>
</author>
<author><name sortKey="Y Ld R M, Tulay" sort="Y Ld R M, Tulay" uniqKey="Y Ld R M T" first="Tülay" last="Y Ld R M">Tülay Y Ld R M</name>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">ISTEX</idno>
<idno type="RBID">ISTEX:FCF50DD73FDEA52D97210EC1DAFEF52F264305C8</idno>
<date when="2005" year="2005">2005</date>
<idno type="doi">10.1007/11539087_13</idno>
<idno type="url">https://api.istex.fr/document/FCF50DD73FDEA52D97210EC1DAFEF52F264305C8/fulltext/pdf</idno>
<idno type="wicri:Area/Istex/Corpus">001A61</idno>
<idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">001A61</idno>
<idno type="wicri:Area/Istex/Curation">001937</idno>
<idno type="wicri:Area/Istex/Checkpoint">000837</idno>
<idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Checkpoint">000837</idno>
<idno type="wicri:doubleKey">0302-9743:2005:Bolat B:active:learning:for</idno>
<idno type="wicri:Area/Main/Merge">000A37</idno>
<idno type="wicri:Area/Main/Curation">000A38</idno>
<idno type="wicri:Area/Main/Exploration">000A38</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Active Learning for Probabilistic Neural Networks</title>
<author><name sortKey="Bolat, Bulent" sort="Bolat, Bulent" uniqKey="Bolat B" first="Bülent" last="Bolat">Bülent Bolat</name>
<affiliation wicri:level="1"><country xml:lang="fr">Turquie</country>
<wicri:regionArea>Dept. of Electronics and Communications Eng., Yildiz Technical University, 34349, Besiktas, Istanbul</wicri:regionArea>
<wicri:noRegion>Istanbul</wicri:noRegion>
</affiliation>
<affiliation wicri:level="1"><country wicri:rule="url">Turquie</country>
</affiliation>
</author>
<author><name sortKey="Y Ld R M, Tulay" sort="Y Ld R M, Tulay" uniqKey="Y Ld R M T" first="Tülay" last="Y Ld R M">Tülay Y Ld R M</name>
<affiliation wicri:level="1"><country xml:lang="fr">Turquie</country>
<wicri:regionArea>Dept. of Electronics and Communications Eng., Yildiz Technical University, 34349, Besiktas, Istanbul</wicri:regionArea>
<wicri:noRegion>Istanbul</wicri:noRegion>
</affiliation>
<affiliation wicri:level="1"><country wicri:rule="url">Turquie</country>
</affiliation>
</author>
</analytic>
<monogr></monogr>
<series><title level="s">Lecture Notes in Computer Science</title>
<imprint><date>2005</date>
</imprint>
<idno type="ISSN">0302-9743</idno>
<idno type="eISSN">1611-3349</idno>
<idno type="ISSN">0302-9743</idno>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><idno type="ISSN">0302-9743</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="Teeft" xml:lang="en"><term>Active learner</term>
<term>Active sampling</term>
<term>Active selection</term>
<term>Algorithm</term>
<term>Basic architecture</term>
<term>Bayesian networks</term>
<term>Better training</term>
<term>Biggest class</term>
<term>Binary</term>
<term>Binary output value</term>
<term>Binary value</term>
<term>Bolat</term>
<term>Breast mass</term>
<term>Bupa</term>
<term>Continuous features</term>
<term>Data collection</term>
<term>Data exchange</term>
<term>Data exchange method</term>
<term>Data replication</term>
<term>Database</term>
<term>Dataset</term>
<term>Datasets</term>
<term>Decision boundary</term>
<term>Decision rule</term>
<term>Digitized image</term>
<term>Entire sample space</term>
<term>Exchange process</term>
<term>Generalization ability</term>
<term>Generalization error</term>
<term>Generalization performance</term>
<term>Glass dataset</term>
<term>Good selection</term>
<term>Hyper planes</term>
<term>Ieee trans</term>
<term>Input layer</term>
<term>Input space</term>
<term>Intelligent systems</term>
<term>International conference</term>
<term>Learner</term>
<term>Less instance numbers</term>
<term>Loss function</term>
<term>Loss functions</term>
<term>Maximum error</term>
<term>Multi layer perceptron</term>
<term>Network toolbox</term>
<term>Neural</term>
<term>Neural information processing systems</term>
<term>Neural network</term>
<term>Neural networks</term>
<term>Neuron</term>
<term>Original location</term>
<term>Other hand</term>
<term>Output layer</term>
<term>Parameter estimation</term>
<term>Parzen windows</term>
<term>Passive observer</term>
<term>Pattern layer neurons</term>
<term>Pattern vector</term>
<term>Probabilistic</term>
<term>Probability density functions</term>
<term>Proc</term>
<term>Redundant instances</term>
<term>Replication</term>
<term>Same training</term>
<term>Sample space</term>
<term>Square error</term>
<term>Summation layer</term>
<term>Support vector machines</term>
<term>Test accuracies</term>
<term>Test accuracy</term>
<term>Training data</term>
<term>Training data boosts</term>
<term>Training examples</term>
<term>Training instance</term>
<term>Training instances</term>
<term>Training pattern</term>
<term>Training patterns</term>
<term>Training sets</term>
<term>Useful training data selection criteria</term>
</keywords>
</textClass>
<langUsage><language ident="en">en</language>
</langUsage>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Abstract: In many neural network applications, the selection of best training set to represent the entire sample space is one of the most important problems. Active learning algorithms in the literature for neural networks are not appropriate for Probabilistic Neural Networks (PNN). In this paper, a new active learning method is proposed for PNN. The method was applied to several benchmark problems.</div>
</front>
</TEI>
<affiliations><list><country><li>Turquie</li>
</country>
</list>
<tree><country name="Turquie"><noRegion><name sortKey="Bolat, Bulent" sort="Bolat, Bulent" uniqKey="Bolat B" first="Bülent" last="Bolat">Bülent Bolat</name>
</noRegion>
<name sortKey="Bolat, Bulent" sort="Bolat, Bulent" uniqKey="Bolat B" first="Bülent" last="Bolat">Bülent Bolat</name>
<name sortKey="Y Ld R M, Tulay" sort="Y Ld R M, Tulay" uniqKey="Y Ld R M T" first="Tülay" last="Y Ld R M">Tülay Y Ld R M</name>
<name sortKey="Y Ld R M, Tulay" sort="Y Ld R M, Tulay" uniqKey="Y Ld R M T" first="Tülay" last="Y Ld R M">Tülay Y Ld R M</name>
</country>
</tree>
</affiliations>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Sarre/explor/MusicSarreV3/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000A38 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000A38 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Wicri/Sarre |area= MusicSarreV3 |flux= Main |étape= Exploration |type= RBID |clé= ISTEX:FCF50DD73FDEA52D97210EC1DAFEF52F264305C8 |texte= Active Learning for Probabilistic Neural Networks }}
This area was generated with Dilib version V0.6.33. |