PVM-based training of large neural architectures
A methodology for parallelizing neural network training algorithms is described, based on the parallel evaluation of the error function and gradient using the parallel virtual machine (PVM). PVM is an integrated set of software tools and libraries that emulates a general-purpose, flexible, heterogen...
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Tagungsbericht |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 2589 vol.4 |
|---|---|
| container_issue | |
| container_start_page | 2584 |
| container_title | |
| container_volume | 4 |
| creator | Plagianakos, V.P. Magoulas, G.D. Nousis, N.K. Vrahatis, M.N. |
| description | A methodology for parallelizing neural network training algorithms is described, based on the parallel evaluation of the error function and gradient using the parallel virtual machine (PVM). PVM is an integrated set of software tools and libraries that emulates a general-purpose, flexible, heterogeneous concurrent computing framework on interconnected computers of various architectures. The methodology proposed has large granularity and low synchronization, and has been implemented and tested. Our results indicate that the relatively easy setup of the PVM (using existing workstations), and parallelization of the training algorithms results in considerable speed-ups especially when large network architectures and training vectors are used. |
| doi_str_mv | 10.1109/IJCNN.2001.938777 |
| format | Conference Proceeding |
| fullrecord | <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_938777</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>938777</ieee_id><sourcerecordid>938777</sourcerecordid><originalsourceid>FETCH-LOGICAL-i147t-df946286e95b21169b2de4f084701e091d1b41eabd8b5b305c6bf6e06989f9ad3</originalsourceid><addsrcrecordid>eNotj0tOwzAUAC0-Em3hALDKBRye4-9boghKUSksgG1lx8_FKATkpAtuD1JZzWKkkYaxSwG1EIDXq4d2s6kbAFGjdNbaIzYTWjsuEZpjNgfrQFpQCk_-BKDjVltzxubj-AFgwCqcMXh-e-TBjxSrqfg85GFXfaWq92VH1UD74vvKl-49T9RN-0LjOTtNvh_p4p8L9np3-9Le8_XTctXerHkWyk48JlSmcYZQh0YIg6GJpBI4ZUEQoIgiKEE-RBd0kKA7E5IhMOgwoY9ywa4O3UxE2--SP3352R5G5S-TTUSi</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>PVM-based training of large neural architectures</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Plagianakos, V.P. ; Magoulas, G.D. ; Nousis, N.K. ; Vrahatis, M.N.</creator><creatorcontrib>Plagianakos, V.P. ; Magoulas, G.D. ; Nousis, N.K. ; Vrahatis, M.N.</creatorcontrib><description>A methodology for parallelizing neural network training algorithms is described, based on the parallel evaluation of the error function and gradient using the parallel virtual machine (PVM). PVM is an integrated set of software tools and libraries that emulates a general-purpose, flexible, heterogeneous concurrent computing framework on interconnected computers of various architectures. The methodology proposed has large granularity and low synchronization, and has been implemented and tested. Our results indicate that the relatively easy setup of the PVM (using existing workstations), and parallelization of the training algorithms results in considerable speed-ups especially when large network architectures and training vectors are used.</description><identifier>ISSN: 1098-7576</identifier><identifier>ISBN: 0780370449</identifier><identifier>ISBN: 9780780370449</identifier><identifier>EISSN: 1558-3902</identifier><identifier>DOI: 10.1109/IJCNN.2001.938777</identifier><language>eng</language><publisher>IEEE</publisher><subject>Artificial intelligence ; Artificial neural networks ; Computer architecture ; Computer errors ; Equations ; Information systems ; Mathematics ; Neurons ; Testing ; Virtual machining</subject><ispartof>IJCNN'01. International Joint Conference on Neural Networks. Proceedings (Cat. No.01CH37222), 2001, Vol.4, p.2584-2589 vol.4</ispartof><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/938777$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,776,780,785,786,2051,4035,4036,27904,54898</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/938777$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Plagianakos, V.P.</creatorcontrib><creatorcontrib>Magoulas, G.D.</creatorcontrib><creatorcontrib>Nousis, N.K.</creatorcontrib><creatorcontrib>Vrahatis, M.N.</creatorcontrib><title>PVM-based training of large neural architectures</title><title>IJCNN'01. International Joint Conference on Neural Networks. Proceedings (Cat. No.01CH37222)</title><addtitle>IJCNN</addtitle><description>A methodology for parallelizing neural network training algorithms is described, based on the parallel evaluation of the error function and gradient using the parallel virtual machine (PVM). PVM is an integrated set of software tools and libraries that emulates a general-purpose, flexible, heterogeneous concurrent computing framework on interconnected computers of various architectures. The methodology proposed has large granularity and low synchronization, and has been implemented and tested. Our results indicate that the relatively easy setup of the PVM (using existing workstations), and parallelization of the training algorithms results in considerable speed-ups especially when large network architectures and training vectors are used.</description><subject>Artificial intelligence</subject><subject>Artificial neural networks</subject><subject>Computer architecture</subject><subject>Computer errors</subject><subject>Equations</subject><subject>Information systems</subject><subject>Mathematics</subject><subject>Neurons</subject><subject>Testing</subject><subject>Virtual machining</subject><issn>1098-7576</issn><issn>1558-3902</issn><isbn>0780370449</isbn><isbn>9780780370449</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2001</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><sourceid>RIE</sourceid><recordid>eNotj0tOwzAUAC0-Em3hALDKBRye4-9boghKUSksgG1lx8_FKATkpAtuD1JZzWKkkYaxSwG1EIDXq4d2s6kbAFGjdNbaIzYTWjsuEZpjNgfrQFpQCk_-BKDjVltzxubj-AFgwCqcMXh-e-TBjxSrqfg85GFXfaWq92VH1UD74vvKl-49T9RN-0LjOTtNvh_p4p8L9np3-9Le8_XTctXerHkWyk48JlSmcYZQh0YIg6GJpBI4ZUEQoIgiKEE-RBd0kKA7E5IhMOgwoY9ywa4O3UxE2--SP3352R5G5S-TTUSi</recordid><startdate>2001</startdate><enddate>2001</enddate><creator>Plagianakos, V.P.</creator><creator>Magoulas, G.D.</creator><creator>Nousis, N.K.</creator><creator>Vrahatis, M.N.</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope></search><sort><creationdate>2001</creationdate><title>PVM-based training of large neural architectures</title><author>Plagianakos, V.P. ; Magoulas, G.D. ; Nousis, N.K. ; Vrahatis, M.N.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i147t-df946286e95b21169b2de4f084701e091d1b41eabd8b5b305c6bf6e06989f9ad3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2001</creationdate><topic>Artificial intelligence</topic><topic>Artificial neural networks</topic><topic>Computer architecture</topic><topic>Computer errors</topic><topic>Equations</topic><topic>Information systems</topic><topic>Mathematics</topic><topic>Neurons</topic><topic>Testing</topic><topic>Virtual machining</topic><toplevel>online_resources</toplevel><creatorcontrib>Plagianakos, V.P.</creatorcontrib><creatorcontrib>Magoulas, G.D.</creatorcontrib><creatorcontrib>Nousis, N.K.</creatorcontrib><creatorcontrib>Vrahatis, M.N.</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Plagianakos, V.P.</au><au>Magoulas, G.D.</au><au>Nousis, N.K.</au><au>Vrahatis, M.N.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>PVM-based training of large neural architectures</atitle><btitle>IJCNN'01. International Joint Conference on Neural Networks. Proceedings (Cat. No.01CH37222)</btitle><stitle>IJCNN</stitle><date>2001</date><risdate>2001</risdate><volume>4</volume><spage>2584</spage><epage>2589 vol.4</epage><pages>2584-2589 vol.4</pages><issn>1098-7576</issn><eissn>1558-3902</eissn><isbn>0780370449</isbn><isbn>9780780370449</isbn><abstract>A methodology for parallelizing neural network training algorithms is described, based on the parallel evaluation of the error function and gradient using the parallel virtual machine (PVM). PVM is an integrated set of software tools and libraries that emulates a general-purpose, flexible, heterogeneous concurrent computing framework on interconnected computers of various architectures. The methodology proposed has large granularity and low synchronization, and has been implemented and tested. Our results indicate that the relatively easy setup of the PVM (using existing workstations), and parallelization of the training algorithms results in considerable speed-ups especially when large network architectures and training vectors are used.</abstract><pub>IEEE</pub><doi>10.1109/IJCNN.2001.938777</doi><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1098-7576 |
| ispartof | IJCNN'01. International Joint Conference on Neural Networks. Proceedings (Cat. No.01CH37222), 2001, Vol.4, p.2584-2589 vol.4 |
| issn | 1098-7576 1558-3902 |
| language | eng |
| recordid | cdi_ieee_primary_938777 |
| source | IEEE Electronic Library (IEL) Conference Proceedings |
| subjects | Artificial intelligence Artificial neural networks Computer architecture Computer errors Equations Information systems Mathematics Neurons Testing Virtual machining |
| title | PVM-based training of large neural architectures |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-27T01%3A13%3A15IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=PVM-based%20training%20of%20large%20neural%20architectures&rft.btitle=IJCNN'01.%20International%20Joint%20Conference%20on%20Neural%20Networks.%20Proceedings%20(Cat.%20No.01CH37222)&rft.au=Plagianakos,%20V.P.&rft.date=2001&rft.volume=4&rft.spage=2584&rft.epage=2589%20vol.4&rft.pages=2584-2589%20vol.4&rft.issn=1098-7576&rft.eissn=1558-3902&rft.isbn=0780370449&rft.isbn_list=9780780370449&rft_id=info:doi/10.1109/IJCNN.2001.938777&rft_dat=%3Cieee_6IE%3E938777%3C/ieee_6IE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=938777&rfr_iscdi=true |