Efficient Domain Decomposition for a Neural Network Learning Algorithm, Used for the Dose Evaluation in External Radiotherapy
Identifieur interne : 003083 ( Main/Exploration ); précédent : 003082; suivant : 003084Efficient Domain Decomposition for a Neural Network Learning Algorithm, Used for the Dose Evaluation in External Radiotherapy
Auteurs : Marc Sauget [France] ; Rémy Laurent [France] ; Julien Henriet [France] ; Michel Salomon [France] ; Régine Gschwind [France] ; Sylvain Contassot-Vivier [France] ; Libor Makovicka [France] ; Charles Soussen [France]Source :
- Lecture Notes in Computer Science [ 0302-9743 ]
Abstract
Abstract: The purpose of this work is to further study the relevance of accelerating the Monte Carlo calculations for the gamma rays external radiotherapy through feed-forward neural networks. We have previously presented a parallel incremental algorithm that builds neural networks of reduced size, while providing high quality approximations of the dose deposit. Our parallel algorithm consists in a regular decomposition of the initial learning dataset (also called learning domain) in as much subsets as available processors. However, the initial learning set presents heterogeneous signal complexities and consequently, the learning times of regular subsets are very different. This paper presents an efficient learning domain decomposition which balances the signal complexities across the processors. As will be shown, the resulting irregular decomposition allows for important gains in learning time of the global network.
Url:
- https://api.istex.fr/ark:/67375/HCB-L6HJG0T5-H/fulltext.pdf
- https://hal.archives-ouvertes.fr/hal-00520154
DOI: 10.1007/978-3-642-15819-3_34
Affiliations:
- France
- Bourgogne-Franche-Comté, Franche-Comté, Grand Est, Lorraine (région)
- Belfort, Montbéliard, Vandœuvre-lès-Nancy
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 000781
- to stream Istex, to step Curation: 000776
- to stream Istex, to step Checkpoint: 000845
- to stream Hal, to step Corpus: 001E71
- to stream Hal, to step Curation: 001E71
- to stream Hal, to step Checkpoint: 002840
- to stream Main, to step Merge: 003140
- to stream Main, to step Curation: 003083
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Efficient Domain Decomposition for a Neural Network Learning Algorithm, Used for the Dose Evaluation in External Radiotherapy</title>
<author><name sortKey="Sauget, Marc" sort="Sauget, Marc" uniqKey="Sauget M" first="Marc" last="Sauget">Marc Sauget</name>
</author>
<author><name sortKey="Laurent, Remy" sort="Laurent, Remy" uniqKey="Laurent R" first="Rémy" last="Laurent">Rémy Laurent</name>
</author>
<author><name sortKey="Henriet, Julien" sort="Henriet, Julien" uniqKey="Henriet J" first="Julien" last="Henriet">Julien Henriet</name>
</author>
<author><name sortKey="Salomon, Michel" sort="Salomon, Michel" uniqKey="Salomon M" first="Michel" last="Salomon">Michel Salomon</name>
</author>
<author><name sortKey="Gschwind, Regine" sort="Gschwind, Regine" uniqKey="Gschwind R" first="Régine" last="Gschwind">Régine Gschwind</name>
</author>
<author><name sortKey="Contassot Vivier, Sylvain" sort="Contassot Vivier, Sylvain" uniqKey="Contassot Vivier S" first="Sylvain" last="Contassot-Vivier">Sylvain Contassot-Vivier</name>
</author>
<author><name sortKey="Makovicka, Libor" sort="Makovicka, Libor" uniqKey="Makovicka L" first="Libor" last="Makovicka">Libor Makovicka</name>
</author>
<author><name sortKey="Soussen, Charles" sort="Soussen, Charles" uniqKey="Soussen C" first="Charles" last="Soussen">Charles Soussen</name>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">ISTEX</idno>
<idno type="RBID">ISTEX:22AC94B35DD35EBDCE365E8E3A2D470E6362D1F0</idno>
<date when="2010" year="2010">2010</date>
<idno type="doi">10.1007/978-3-642-15819-3_34</idno>
<idno type="url">https://api.istex.fr/ark:/67375/HCB-L6HJG0T5-H/fulltext.pdf</idno>
<idno type="wicri:Area/Istex/Corpus">000781</idno>
<idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">000781</idno>
<idno type="wicri:Area/Istex/Curation">000776</idno>
<idno type="wicri:Area/Istex/Checkpoint">000845</idno>
<idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Checkpoint">000845</idno>
<idno type="wicri:doubleKey">0302-9743:2010:Sauget M:efficient:domain:decomposition</idno>
<idno type="wicri:source">HAL</idno>
<idno type="RBID">Hal:hal-00520154</idno>
<idno type="url">https://hal.archives-ouvertes.fr/hal-00520154</idno>
<idno type="wicri:Area/Hal/Corpus">001E71</idno>
<idno type="wicri:Area/Hal/Curation">001E71</idno>
<idno type="wicri:Area/Hal/Checkpoint">002840</idno>
<idno type="wicri:explorRef" wicri:stream="Hal" wicri:step="Checkpoint">002840</idno>
<idno type="wicri:Area/Main/Merge">003140</idno>
<idno type="wicri:Area/Main/Curation">003083</idno>
<idno type="wicri:Area/Main/Exploration">003083</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Efficient Domain Decomposition for a Neural Network Learning Algorithm, Used for the Dose Evaluation in External Radiotherapy</title>
<author><name sortKey="Sauget, Marc" sort="Sauget, Marc" uniqKey="Sauget M" first="Marc" last="Sauget">Marc Sauget</name>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>Femto-ST, ENISYS/IRMA, F-25210, Montbéliard</wicri:regionArea>
<placeName><region type="region" nuts="2">Bourgogne-Franche-Comté</region>
<region type="old region" nuts="2">Franche-Comté</region>
<settlement type="city">Montbéliard</settlement>
</placeName>
</affiliation>
<affiliation wicri:level="1"><country wicri:rule="url">France</country>
</affiliation>
</author>
<author><name sortKey="Laurent, Remy" sort="Laurent, Remy" uniqKey="Laurent R" first="Rémy" last="Laurent">Rémy Laurent</name>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>Femto-ST, ENISYS/IRMA, F-25210, Montbéliard</wicri:regionArea>
<placeName><region type="region" nuts="2">Bourgogne-Franche-Comté</region>
<region type="old region" nuts="2">Franche-Comté</region>
<settlement type="city">Montbéliard</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Henriet, Julien" sort="Henriet, Julien" uniqKey="Henriet J" first="Julien" last="Henriet">Julien Henriet</name>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>Femto-ST, ENISYS/IRMA, F-25210, Montbéliard</wicri:regionArea>
<placeName><region type="region" nuts="2">Bourgogne-Franche-Comté</region>
<region type="old region" nuts="2">Franche-Comté</region>
<settlement type="city">Montbéliard</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Salomon, Michel" sort="Salomon, Michel" uniqKey="Salomon M" first="Michel" last="Salomon">Michel Salomon</name>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>LIFC/AND, University of Franche-Comté, F-90000, Belfort</wicri:regionArea>
<placeName><region type="region" nuts="2">Bourgogne-Franche-Comté</region>
<region type="old region" nuts="2">Franche-Comté</region>
<settlement type="city">Belfort</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Gschwind, Regine" sort="Gschwind, Regine" uniqKey="Gschwind R" first="Régine" last="Gschwind">Régine Gschwind</name>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>Femto-ST, ENISYS/IRMA, F-25210, Montbéliard</wicri:regionArea>
<placeName><region type="region" nuts="2">Bourgogne-Franche-Comté</region>
<region type="old region" nuts="2">Franche-Comté</region>
<settlement type="city">Montbéliard</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Contassot Vivier, Sylvain" sort="Contassot Vivier, Sylvain" uniqKey="Contassot Vivier S" first="Sylvain" last="Contassot-Vivier">Sylvain Contassot-Vivier</name>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>LORIA, University of Nancy, Campus Scientifique, BP 239, F-54506, Vandoeuvre-lès-Nancy Cedex</wicri:regionArea>
<placeName><region type="region" nuts="2">Grand Est</region>
<region type="old region" nuts="2">Lorraine (région)</region>
<settlement type="city">Vandœuvre-lès-Nancy</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Makovicka, Libor" sort="Makovicka, Libor" uniqKey="Makovicka L" first="Libor" last="Makovicka">Libor Makovicka</name>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>Femto-ST, ENISYS/IRMA, F-25210, Montbéliard</wicri:regionArea>
<placeName><region type="region" nuts="2">Bourgogne-Franche-Comté</region>
<region type="old region" nuts="2">Franche-Comté</region>
<settlement type="city">Montbéliard</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Soussen, Charles" sort="Soussen, Charles" uniqKey="Soussen C" first="Charles" last="Soussen">Charles Soussen</name>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>Faculté des sciences et techniques, CRAN, F-54506, Vandoeuvre-lès-Nancy Cedex</wicri:regionArea>
<placeName><region type="region" nuts="2">Grand Est</region>
<region type="old region" nuts="2">Lorraine (région)</region>
<settlement type="city">Vandœuvre-lès-Nancy</settlement>
</placeName>
</affiliation>
</author>
</analytic>
<monogr></monogr>
<series><title level="s" type="main" xml:lang="en">Lecture Notes in Computer Science</title>
<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></textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Abstract: The purpose of this work is to further study the relevance of accelerating the Monte Carlo calculations for the gamma rays external radiotherapy through feed-forward neural networks. We have previously presented a parallel incremental algorithm that builds neural networks of reduced size, while providing high quality approximations of the dose deposit. Our parallel algorithm consists in a regular decomposition of the initial learning dataset (also called learning domain) in as much subsets as available processors. However, the initial learning set presents heterogeneous signal complexities and consequently, the learning times of regular subsets are very different. This paper presents an efficient learning domain decomposition which balances the signal complexities across the processors. As will be shown, the resulting irregular decomposition allows for important gains in learning time of the global network.</div>
</front>
</TEI>
<affiliations><list><country><li>France</li>
</country>
<region><li>Bourgogne-Franche-Comté</li>
<li>Franche-Comté</li>
<li>Grand Est</li>
<li>Lorraine (région)</li>
</region>
<settlement><li>Belfort</li>
<li>Montbéliard</li>
<li>Vandœuvre-lès-Nancy</li>
</settlement>
</list>
<tree><country name="France"><region name="Bourgogne-Franche-Comté"><name sortKey="Sauget, Marc" sort="Sauget, Marc" uniqKey="Sauget M" first="Marc" last="Sauget">Marc Sauget</name>
</region>
<name sortKey="Contassot Vivier, Sylvain" sort="Contassot Vivier, Sylvain" uniqKey="Contassot Vivier S" first="Sylvain" last="Contassot-Vivier">Sylvain Contassot-Vivier</name>
<name sortKey="Gschwind, Regine" sort="Gschwind, Regine" uniqKey="Gschwind R" first="Régine" last="Gschwind">Régine Gschwind</name>
<name sortKey="Henriet, Julien" sort="Henriet, Julien" uniqKey="Henriet J" first="Julien" last="Henriet">Julien Henriet</name>
<name sortKey="Laurent, Remy" sort="Laurent, Remy" uniqKey="Laurent R" first="Rémy" last="Laurent">Rémy Laurent</name>
<name sortKey="Makovicka, Libor" sort="Makovicka, Libor" uniqKey="Makovicka L" first="Libor" last="Makovicka">Libor Makovicka</name>
<name sortKey="Salomon, Michel" sort="Salomon, Michel" uniqKey="Salomon M" first="Michel" last="Salomon">Michel Salomon</name>
<name sortKey="Sauget, Marc" sort="Sauget, Marc" uniqKey="Sauget M" first="Marc" last="Sauget">Marc Sauget</name>
<name sortKey="Soussen, Charles" sort="Soussen, Charles" uniqKey="Soussen C" first="Charles" last="Soussen">Charles Soussen</name>
</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 003083 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 003083 | 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= ISTEX:22AC94B35DD35EBDCE365E8E3A2D470E6362D1F0 |texte= Efficient Domain Decomposition for a Neural Network Learning Algorithm, Used for the Dose Evaluation in External Radiotherapy }}
This area was generated with Dilib version V0.6.33. |