Rejection strategy for Convolutional Neural Network by adaptive topology applied to handwritten digits recognition
Identifieur interne : 005C78 ( Main/Exploration ); précédent : 005C77; suivant : 005C79Rejection strategy for Convolutional Neural Network by adaptive topology applied to handwritten digits recognition
Auteurs : Hubert Cecotti ; Abdel Belaïd [France]Source :
English descriptors
Abstract
In this paper, we propose a rejection strategy for convolutional neural network models. The purpose of this work is to adapt the network's topology in function of the geometrical error. A self-organizing map is used to change the links between the layers leading to a geometric image transformation occurring directly inside the network. Instead of learning all the possible deformation of a pattern, ambiguous patterns are rejected and the network's topology is modified in function of their geometric errors thanks to a specialized self-organizing map. Our objective is to show how an adaptive topology, without a new learning, can improve the recognition of rejected patterns in the case of handwritten digits.
Affiliations:
- France
- Grand Est, Lorraine (région)
- Nancy
- Centre national de la recherche scientifique, Institut national de recherche en informatique et en automatique, Laboratoire lorrain de recherche en informatique et ses applications, Université de Lorraine
Links toward previous steps (curation, corpus...)
- to stream Crin, to step Corpus: 004160
- to stream Crin, to step Curation: 004160
- to stream Crin, to step Checkpoint: 000126
- to stream Main, to step Merge: 005F01
- to stream Main, to step Curation: 005C78
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" wicri:score="799">Rejection strategy for Convolutional Neural Network by adaptive topology applied to handwritten digits recognition</title></titleStmt><publicationStmt><idno type="RBID">CRIN:cecotti05a</idno><date when="2005" year="2005">2005</date><idno type="wicri:Area/Crin/Corpus">004160</idno><idno type="wicri:Area/Crin/Curation">004160</idno><idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Curation">004160</idno><idno type="wicri:Area/Crin/Checkpoint">000126</idno><idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Checkpoint">000126</idno><idno type="wicri:Area/Main/Merge">005F01</idno><idno type="wicri:Area/Main/Curation">005C78</idno><idno type="wicri:Area/Main/Exploration">005C78</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Rejection strategy for Convolutional Neural Network by adaptive topology applied to handwritten digits recognition</title><author><name sortKey="Cecotti, Hubert" sort="Cecotti, Hubert" uniqKey="Cecotti H" first="Hubert" last="Cecotti">Hubert Cecotti</name></author><author><name sortKey="Belaid, Abdel" sort="Belaid, Abdel" uniqKey="Belaid A" first="Abdel" last="Belaid">Abdel Belaïd</name><affiliation><country>France</country><placeName><settlement type="city">Nancy</settlement><region type="region" nuts="2">Grand Est</region><region type="region" nuts="2">Lorraine (région)</region></placeName><orgName type="laboratoire" n="5">Laboratoire lorrain de recherche en informatique et ses applications</orgName><orgName type="university">Université de Lorraine</orgName><orgName type="institution">Centre national de la recherche scientifique</orgName><orgName type="institution">Institut national de recherche en informatique et en automatique</orgName></affiliation></author></analytic></biblStruct></sourceDesc></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>adaptive topology</term><term>character recognition</term><term>convolutional neural network</term><term>rejection strategy</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en" wicri:score="3157">In this paper, we propose a rejection strategy for convolutional neural network models. The purpose of this work is to adapt the network's topology in function of the geometrical error. A self-organizing map is used to change the links between the layers leading to a geometric image transformation occurring directly inside the network. Instead of learning all the possible deformation of a pattern, ambiguous patterns are rejected and the network's topology is modified in function of their geometric errors thanks to a specialized self-organizing map. Our objective is to show how an adaptive topology, without a new learning, can improve the recognition of rejected patterns in the case of handwritten digits.</div></front></TEI><affiliations><list><country><li>France</li></country><region><li>Grand Est</li><li>Lorraine (région)</li></region><settlement><li>Nancy</li></settlement><orgName><li>Centre national de la recherche scientifique</li><li>Institut national de recherche en informatique et en automatique</li><li>Laboratoire lorrain de recherche en informatique et ses applications</li><li>Université de Lorraine</li></orgName></list><tree><noCountry><name sortKey="Cecotti, Hubert" sort="Cecotti, Hubert" uniqKey="Cecotti H" first="Hubert" last="Cecotti">Hubert Cecotti</name></noCountry><country name="France"><region name="Grand Est"><name sortKey="Belaid, Abdel" sort="Belaid, Abdel" uniqKey="Belaid A" first="Abdel" last="Belaid">Abdel Belaïd</name></region></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 005C78 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 005C78 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Lorraine
|area= InforLorV4
|flux= Main
|étape= Exploration
|type= RBID
|clé= CRIN:cecotti05a
|texte= Rejection strategy for Convolutional Neural Network by adaptive topology applied to handwritten digits recognition
}}
| This area was generated with Dilib version V0.6.33. |  |