Digital hardware implementation of 2D compatible neural networks
Identifieur interne : 001A16 ( Crin/Checkpoint ); précédent : 001A15; suivant : 001A17Digital hardware implementation of 2D compatible neural networks
Auteurs : Bernard GirauSource :
English descriptors
- KwdEn :
Abstract
The work described in this paper aims at developing neural architectures that are easy to map onto FPGAs, thanks to a simplified topology and an original data exchange scheme, without significant loss of approximation capability. It has been achieved thanks to the definition of a set of neural models called Field Programmable Neural Arrays (FPNA). FPNAs may lead to the definition of neural networks adapted to hardware topological constraints. Different such neural networks may be derived from a given FPNA. They are called Field Programmed Neural Networks (FPNN). They reconcile the high connection density of neural architectures with the need of a limited interconnection scheme in hardware implementations. This paper focuses on the definition and implementation of FPNN parallel computation.
Links toward previous steps (curation, corpus...)
Links to Exploration step
CRIN:girau00rLe document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" wicri:score="361">Digital hardware implementation of 2D compatible neural networks</title>
</titleStmt>
<publicationStmt><idno type="RBID">CRIN:girau00r</idno>
<date when="2000" year="2000">2000</date>
<idno type="wicri:Area/Crin/Corpus">002B25</idno>
<idno type="wicri:Area/Crin/Curation">002B25</idno>
<idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Curation">002B25</idno>
<idno type="wicri:Area/Crin/Checkpoint">001A16</idno>
<idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Checkpoint">001A16</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en">Digital hardware implementation of 2D compatible neural networks</title>
<author><name sortKey="Girau, Bernard" sort="Girau, Bernard" uniqKey="Girau B" first="Bernard" last="Girau">Bernard Girau</name>
</author>
</analytic>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>digital implementation</term>
<term>fpgas</term>
<term>neural networks</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en" wicri:score="3208">The work described in this paper aims at developing neural architectures that are easy to map onto FPGAs, thanks to a simplified topology and an original data exchange scheme, without significant loss of approximation capability. It has been achieved thanks to the definition of a set of neural models called Field Programmable Neural Arrays (FPNA). FPNAs may lead to the definition of neural networks adapted to hardware topological constraints. Different such neural networks may be derived from a given FPNA. They are called Field Programmed Neural Networks (FPNN). They reconcile the high connection density of neural architectures with the need of a limited interconnection scheme in hardware implementations. This paper focuses on the definition and implementation of FPNN parallel computation.</div>
</front>
</TEI>
<BibTex type="inproceedings"><ref>girau00r</ref>
<crinnumber>A00-R-292</crinnumber>
<category>3</category>
<equipe>CORTEX</equipe>
<author><e>Girau, Bernard</e>
</author>
<title>Digital hardware implementation of 2D compatible neural networks</title>
<booktitle>{International Joint Conference on Neural Networks}</booktitle>
<year>2000</year>
<month>Jul</month>
<keywords><e>neural networks</e>
<e>digital implementation</e>
<e>fpgas</e>
</keywords>
<abstract>The work described in this paper aims at developing neural architectures that are easy to map onto FPGAs, thanks to a simplified topology and an original data exchange scheme, without significant loss of approximation capability. It has been achieved thanks to the definition of a set of neural models called Field Programmable Neural Arrays (FPNA). FPNAs may lead to the definition of neural networks adapted to hardware topological constraints. Different such neural networks may be derived from a given FPNA. They are called Field Programmed Neural Networks (FPNN). They reconcile the high connection density of neural architectures with the need of a limited interconnection scheme in hardware implementations. This paper focuses on the definition and implementation of FPNN parallel computation.</abstract>
</BibTex>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Crin/Checkpoint
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 001A16 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Crin/Checkpoint/biblio.hfd -nk 001A16 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Wicri/Lorraine |area= InforLorV4 |flux= Crin |étape= Checkpoint |type= RBID |clé= CRIN:girau00r |texte= Digital hardware implementation of 2D compatible neural networks }}
This area was generated with Dilib version V0.6.33. |