Competition in high dimensional spaces using a sparse approximation of neural fields.
Identifieur interne : 000097 ( PubMed/Corpus ); précédent : 000096; suivant : 000098Competition in high dimensional spaces using a sparse approximation of neural fields.
Auteurs : Jean-Charles Quinton ; Bernard Girau ; Mathieu LefortSource :
- Advances in experimental medicine and biology [ 0065-2598 ] ; 2011.
English descriptors
- KwdEn :
- MESH :
Abstract
The Continuum Neural Field Theory implements competition within topologically organized neural networks with lateral inhibitory connections. However, due to the polynomial complexity of matrix-based implementations, updating dense representations of the activity becomes computationally intractable when an adaptive resolution or an arbitrary number of input dimensions is required. This paper proposes an alternative to self-organizing maps with a sparse implementation based on Gaussian mixture models, promoting a trade-off in redundancy for higher computational efficiency and alleviating constraints on the underlying substrate.This version reproduces the emergent attentional properties of the original equations, by directly applying them within a continuous approximation of a high dimensional neural field. The model is compatible with preprocessed sensory flows but can also be interfaced with artificial systems. This is particularly important for sensorimotor systems, where decisions and motor actions must be taken and updated in real-time. Preliminary tests are performed on a reactive color tracking application, using spatially distributed color features.
DOI: 10.1007/978-1-4614-0164-3_11
PubMed: 21744215
Links to Exploration step
pubmed:21744215Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Competition in high dimensional spaces using a sparse approximation of neural fields.</title><author><name sortKey="Quinton, Jean Charles" sort="Quinton, Jean Charles" uniqKey="Quinton J" first="Jean-Charles" last="Quinton">Jean-Charles Quinton</name><affiliation><nlm:affiliation>INRIA/LORIA Laboratory, Campus Scientifique, B.P. 239, 54506 Vandoeuvre-lès-Nancy Cedex, France. quintonj@loria.fr</nlm:affiliation></affiliation></author><author><name sortKey="Girau, Bernard" sort="Girau, Bernard" uniqKey="Girau B" first="Bernard" last="Girau">Bernard Girau</name></author><author><name sortKey="Lefort, Mathieu" sort="Lefort, Mathieu" uniqKey="Lefort M" first="Mathieu" last="Lefort">Mathieu Lefort</name></author></titleStmt><publicationStmt><idno type="wicri:source">PubMed</idno><date when="2011">2011</date><idno type="doi">10.1007/978-1-4614-0164-3_11</idno><idno type="RBID">pubmed:21744215</idno><idno type="pmid">21744215</idno><idno type="wicri:Area/PubMed/Corpus">000097</idno><idno type="wicri:explorRef" wicri:stream="PubMed" wicri:step="Corpus" wicri:corpus="PubMed">000097</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Competition in high dimensional spaces using a sparse approximation of neural fields.</title><author><name sortKey="Quinton, Jean Charles" sort="Quinton, Jean Charles" uniqKey="Quinton J" first="Jean-Charles" last="Quinton">Jean-Charles Quinton</name><affiliation><nlm:affiliation>INRIA/LORIA Laboratory, Campus Scientifique, B.P. 239, 54506 Vandoeuvre-lès-Nancy Cedex, France. quintonj@loria.fr</nlm:affiliation></affiliation></author><author><name sortKey="Girau, Bernard" sort="Girau, Bernard" uniqKey="Girau B" first="Bernard" last="Girau">Bernard Girau</name></author><author><name sortKey="Lefort, Mathieu" sort="Lefort, Mathieu" uniqKey="Lefort M" first="Mathieu" last="Lefort">Mathieu Lefort</name></author></analytic><series><title level="j">Advances in experimental medicine and biology</title><idno type="ISSN">0065-2598</idno><imprint><date when="2011" type="published">2011</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Models, Theoretical</term><term>Nerve Net</term><term>Neural Networks (Computer)</term></keywords><keywords scheme="MESH" xml:lang="en"><term>Models, Theoretical</term><term>Nerve Net</term><term>Neural Networks (Computer)</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">The Continuum Neural Field Theory implements competition within topologically organized neural networks with lateral inhibitory connections. However, due to the polynomial complexity of matrix-based implementations, updating dense representations of the activity becomes computationally intractable when an adaptive resolution or an arbitrary number of input dimensions is required. This paper proposes an alternative to self-organizing maps with a sparse implementation based on Gaussian mixture models, promoting a trade-off in redundancy for higher computational efficiency and alleviating constraints on the underlying substrate.This version reproduces the emergent attentional properties of the original equations, by directly applying them within a continuous approximation of a high dimensional neural field. The model is compatible with preprocessed sensory flows but can also be interfaced with artificial systems. This is particularly important for sensorimotor systems, where decisions and motor actions must be taken and updated in real-time. Preliminary tests are performed on a reactive color tracking application, using spatially distributed color features.</div></front></TEI><pubmed><MedlineCitation Owner="NLM" Status="MEDLINE"><PMID Version="1">21744215</PMID><DateCreated><Year>2011</Year><Month>07</Month><Day>11</Day></DateCreated><DateCompleted><Year>2011</Year><Month>11</Month><Day>29</Day></DateCompleted><Article PubModel="Print"><Journal><ISSN IssnType="Print">0065-2598</ISSN><JournalIssue CitedMedium="Print"><Volume>718</Volume><PubDate><Year>2011</Year></PubDate></JournalIssue><Title>Advances in experimental medicine and biology</Title><ISOAbbreviation>Adv. Exp. Med. Biol.</ISOAbbreviation></Journal><ArticleTitle>Competition in high dimensional spaces using a sparse approximation of neural fields.</ArticleTitle><Pagination><MedlinePgn>123-37</MedlinePgn></Pagination><ELocationID EIdType="doi" ValidYN="Y">10.1007/978-1-4614-0164-3_11</ELocationID><Abstract><AbstractText>The Continuum Neural Field Theory implements competition within topologically organized neural networks with lateral inhibitory connections. However, due to the polynomial complexity of matrix-based implementations, updating dense representations of the activity becomes computationally intractable when an adaptive resolution or an arbitrary number of input dimensions is required. This paper proposes an alternative to self-organizing maps with a sparse implementation based on Gaussian mixture models, promoting a trade-off in redundancy for higher computational efficiency and alleviating constraints on the underlying substrate.This version reproduces the emergent attentional properties of the original equations, by directly applying them within a continuous approximation of a high dimensional neural field. The model is compatible with preprocessed sensory flows but can also be interfaced with artificial systems. This is particularly important for sensorimotor systems, where decisions and motor actions must be taken and updated in real-time. Preliminary tests are performed on a reactive color tracking application, using spatially distributed color features.</AbstractText></Abstract><AuthorList CompleteYN="Y"><Author ValidYN="Y"><LastName>Quinton</LastName><ForeName>Jean-Charles</ForeName><Initials>JC</Initials><AffiliationInfo><Affiliation>INRIA/LORIA Laboratory, Campus Scientifique, B.P. 239, 54506 Vandoeuvre-lès-Nancy Cedex, France. quintonj@loria.fr</Affiliation></AffiliationInfo></Author><Author ValidYN="Y"><LastName>Girau</LastName><ForeName>Bernard</ForeName><Initials>B</Initials></Author><Author ValidYN="Y"><LastName>Lefort</LastName><ForeName>Mathieu</ForeName><Initials>M</Initials></Author></AuthorList><Language>eng</Language><PublicationTypeList><PublicationType UI="D016428">Journal Article</PublicationType></PublicationTypeList></Article><MedlineJournalInfo><Country>United States</Country><MedlineTA>Adv Exp Med Biol</MedlineTA><NlmUniqueID>0121103</NlmUniqueID><ISSNLinking>0065-2598</ISSNLinking></MedlineJournalInfo><CitationSubset>IM</CitationSubset><MeshHeadingList><MeshHeading><DescriptorName MajorTopicYN="Y" UI="D008962">Models, Theoretical</DescriptorName></MeshHeading><MeshHeading><DescriptorName MajorTopicYN="Y" UI="D009415">Nerve Net</DescriptorName></MeshHeading><MeshHeading><DescriptorName MajorTopicYN="Y" UI="D016571">Neural Networks (Computer)</DescriptorName></MeshHeading></MeshHeadingList></MedlineCitation><PubmedData><History><PubMedPubDate PubStatus="entrez"><Year>2011</Year><Month>7</Month><Day>12</Day><Hour>6</Hour><Minute>0</Minute></PubMedPubDate><PubMedPubDate PubStatus="pubmed"><Year>2011</Year><Month>7</Month><Day>12</Day><Hour>6</Hour><Minute>0</Minute></PubMedPubDate><PubMedPubDate PubStatus="medline"><Year>2011</Year><Month>12</Month><Day>13</Day><Hour>0</Hour><Minute>0</Minute></PubMedPubDate></History><PublicationStatus>ppublish</PublicationStatus><ArticleIdList><ArticleId IdType="doi">10.1007/978-1-4614-0164-3_11</ArticleId><ArticleId IdType="pubmed">21744215</ArticleId></ArticleIdList></PubmedData></pubmed></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/PubMed/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000097 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/PubMed/Corpus/biblio.hfd -nk 000097 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Lorraine
|area= InforLorV4
|flux= PubMed
|étape= Corpus
|type= RBID
|clé= pubmed:21744215
|texte= Competition in high dimensional spaces using a sparse approximation of neural fields.
}}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/PubMed/Corpus/RBID.i -Sk "pubmed:21744215" \
| HfdSelect -Kh $EXPLOR_AREA/Data/PubMed/Corpus/biblio.hfd \
| NlmPubMed2Wicri -a InforLorV4
| This area was generated with Dilib version V0.6.33. |  |