Online Symbolic-Sequence Prediction with Discrete-Time Recurrent Neural Networks

Online Symbolic-Sequence Prediction with Discrete-Time Recurrent Neural Networks

This paper studies the use of discrete-time recurrent neural networks for predicting the next symbol in a sequence. The focus is on online prediction, a task much harder than the classical o.ine grammatical inference with neural networks. The results obtained show that the performance of recurrent n...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Pérez-Ortiz, Juan Antonio, Calera-Rubio, Jorge, Forcada, Mikel L.
Format: Buchkapitel
Sprache:eng
Schlagworte:
Applied sciences
Artificial intelligence
Computer science
control theory
systems
Connectionism. Neural networks
Exact sciences and technology
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 724
container_issue
container_start_page 719
container_title
container_volume 2130
creator Pérez-Ortiz, Juan Antonio
Calera-Rubio, Jorge
Forcada, Mikel L.
description This paper studies the use of discrete-time recurrent neural networks for predicting the next symbol in a sequence. The focus is on online prediction, a task much harder than the classical o.ine grammatical inference with neural networks. The results obtained show that the performance of recurrent networks working online is acceptable when sequences come from finite-state machines or even from some chaotic sources. When predicting texts in human language, however, dynamics seem to be too complex to be correctly learned in real-time by the net. Two algorithms are considered for network training: real-time recurrent learning and the decoupled extended Kalman filter.
doi_str_mv 10.1007/3-540-44668-0_100
format Book Chapter
fullrecord <record><control><sourceid>proquest_pasca</sourceid><recordid>TN_cdi_pascalfrancis_primary_14045253</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>EBC3073004_106_721</sourcerecordid><originalsourceid>FETCH-LOGICAL-p313t-ab7b35ccb4c49d7c5ca825abd750d37197af5dfcf3f3ce0238a30de45345e9663</originalsourceid><addsrcrecordid>eNotkEtPwzAQhM1TlMIP4JYLR4PttePkiHhLFVS0nC3H2YAhTYqdquq_x5TuZaTZnZXmI-SCsyvOmL4GqiSjUuZ5QZlJ1h45hWRtHbZPRjznnALI8mC3ELLI1SEZMWCCllrCMRmVqlBCAhQn5DzGL5YGBJeiGJHpa9f6DrPZZlH1rXd0hj8r7Bxm04C1d4Pvu2zth8_szkcXcEA69wvM3tCtQsBuyF5wFWybZFj34TuekaPGthHPdzom7w_389snOnl9fL69mdAlcBiorXQFyrlKOlnW2ilnC6FsVWvFatC81LZRdeMaaMAhE1BYYDVKBVJhmecwJpf_f5c2Ots2wXbOR7MMfmHDxvCEQgkF6e76_y6mVfeBwVR9_x0Tyz-c2oBJ1MyWp9laKSF3n0OfWMTB4F_EpbKpqPu0ywFDNMA0MCZTJDdacPgFhj16EQ</addsrcrecordid><sourcetype>Index Database</sourcetype><iscdi>true</iscdi><recordtype>book_chapter</recordtype><pqid>EBC3073004_106_721</pqid></control><display><type>book_chapter</type><title>Online Symbolic-Sequence Prediction with Discrete-Time Recurrent Neural Networks</title><source>Springer Books</source><creator>Pérez-Ortiz, Juan Antonio ; Calera-Rubio, Jorge ; Forcada, Mikel L.</creator><contributor>Dorffner, Georg ; Bischof, Horst ; Hornik, Kurt ; Hornik, Kurt ; Dorffner, Georg ; Bischof, Horst</contributor><creatorcontrib>Pérez-Ortiz, Juan Antonio ; Calera-Rubio, Jorge ; Forcada, Mikel L. ; Dorffner, Georg ; Bischof, Horst ; Hornik, Kurt ; Hornik, Kurt ; Dorffner, Georg ; Bischof, Horst</creatorcontrib><description>This paper studies the use of discrete-time recurrent neural networks for predicting the next symbol in a sequence. The focus is on online prediction, a task much harder than the classical o.ine grammatical inference with neural networks. The results obtained show that the performance of recurrent networks working online is acceptable when sequences come from finite-state machines or even from some chaotic sources. When predicting texts in human language, however, dynamics seem to be too complex to be correctly learned in real-time by the net. Two algorithms are considered for network training: real-time recurrent learning and the decoupled extended Kalman filter.</description><identifier>ISSN: 0302-9743</identifier><identifier>ISBN: 3540424865</identifier><identifier>ISBN: 9783540424864</identifier><identifier>EISSN: 1611-3349</identifier><identifier>EISBN: 3540446680</identifier><identifier>EISBN: 9783540446682</identifier><identifier>DOI: 10.1007/3-540-44668-0_100</identifier><identifier>OCLC: 958524338</identifier><identifier>LCCallNum: Q334-342</identifier><language>eng</language><publisher>Germany: Springer Berlin / Heidelberg</publisher><subject>Applied sciences ; Artificial intelligence ; Computer science; control theory; systems ; Connectionism. Neural networks ; Exact sciences and technology</subject><ispartof>Artificial Neural Networks - ICANN 2001, 2001, Vol.2130, p.719-724</ispartof><rights>Springer-Verlag Berlin Heidelberg 2001</rights><rights>2002 INIST-CNRS</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><relation>Lecture Notes in Computer Science</relation></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Uhttps://ebookcentral.proquest.com/covers/3073004-l.jpg</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/3-540-44668-0_100$$EPDF$$P50$$Gspringer$$H</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/3-540-44668-0_100$$EHTML$$P50$$Gspringer$$H</linktohtml><link.rule.ids>309,310,779,780,784,789,790,793,4050,4051,27925,38255,41442,42511</link.rule.ids><backlink>$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&amp;idt=14045253$$DView record in Pascal Francis$$Hfree_for_read</backlink></links><search><contributor>Dorffner, Georg</contributor><contributor>Bischof, Horst</contributor><contributor>Hornik, Kurt</contributor><contributor>Hornik, Kurt</contributor><contributor>Dorffner, Georg</contributor><contributor>Bischof, Horst</contributor><creatorcontrib>Pérez-Ortiz, Juan Antonio</creatorcontrib><creatorcontrib>Calera-Rubio, Jorge</creatorcontrib><creatorcontrib>Forcada, Mikel L.</creatorcontrib><title>Online Symbolic-Sequence Prediction with Discrete-Time Recurrent Neural Networks</title><title>Artificial Neural Networks - ICANN 2001</title><description>This paper studies the use of discrete-time recurrent neural networks for predicting the next symbol in a sequence. The focus is on online prediction, a task much harder than the classical o.ine grammatical inference with neural networks. The results obtained show that the performance of recurrent networks working online is acceptable when sequences come from finite-state machines or even from some chaotic sources. When predicting texts in human language, however, dynamics seem to be too complex to be correctly learned in real-time by the net. Two algorithms are considered for network training: real-time recurrent learning and the decoupled extended Kalman filter.</description><subject>Applied sciences</subject><subject>Artificial intelligence</subject><subject>Computer science; control theory; systems</subject><subject>Connectionism. Neural networks</subject><subject>Exact sciences and technology</subject><issn>0302-9743</issn><issn>1611-3349</issn><isbn>3540424865</isbn><isbn>9783540424864</isbn><isbn>3540446680</isbn><isbn>9783540446682</isbn><fulltext>true</fulltext><rsrctype>book_chapter</rsrctype><creationdate>2001</creationdate><recordtype>book_chapter</recordtype><recordid>eNotkEtPwzAQhM1TlMIP4JYLR4PttePkiHhLFVS0nC3H2YAhTYqdquq_x5TuZaTZnZXmI-SCsyvOmL4GqiSjUuZ5QZlJ1h45hWRtHbZPRjznnALI8mC3ELLI1SEZMWCCllrCMRmVqlBCAhQn5DzGL5YGBJeiGJHpa9f6DrPZZlH1rXd0hj8r7Bxm04C1d4Pvu2zth8_szkcXcEA69wvM3tCtQsBuyF5wFWybZFj34TuekaPGthHPdzom7w_389snOnl9fL69mdAlcBiorXQFyrlKOlnW2ilnC6FsVWvFatC81LZRdeMaaMAhE1BYYDVKBVJhmecwJpf_f5c2Ots2wXbOR7MMfmHDxvCEQgkF6e76_y6mVfeBwVR9_x0Tyz-c2oBJ1MyWp9laKSF3n0OfWMTB4F_EpbKpqPu0ywFDNMA0MCZTJDdacPgFhj16EQ</recordid><startdate>2001</startdate><enddate>2001</enddate><creator>Pérez-Ortiz, Juan Antonio</creator><creator>Calera-Rubio, Jorge</creator><creator>Forcada, Mikel L.</creator><general>Springer Berlin / Heidelberg</general><general>Springer Berlin Heidelberg</general><general>Springer</general><scope>FFUUA</scope><scope>IQODW</scope></search><sort><creationdate>2001</creationdate><title>Online Symbolic-Sequence Prediction with Discrete-Time Recurrent Neural Networks</title><author>Pérez-Ortiz, Juan Antonio ; Calera-Rubio, Jorge ; Forcada, Mikel L.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-p313t-ab7b35ccb4c49d7c5ca825abd750d37197af5dfcf3f3ce0238a30de45345e9663</frbrgroupid><rsrctype>book_chapters</rsrctype><prefilter>book_chapters</prefilter><language>eng</language><creationdate>2001</creationdate><topic>Applied sciences</topic><topic>Artificial intelligence</topic><topic>Computer science; control theory; systems</topic><topic>Connectionism. Neural networks</topic><topic>Exact sciences and technology</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Pérez-Ortiz, Juan Antonio</creatorcontrib><creatorcontrib>Calera-Rubio, Jorge</creatorcontrib><creatorcontrib>Forcada, Mikel L.</creatorcontrib><collection>ProQuest Ebook Central - Book Chapters - Demo use only</collection><collection>Pascal-Francis</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Pérez-Ortiz, Juan Antonio</au><au>Calera-Rubio, Jorge</au><au>Forcada, Mikel L.</au><au>Dorffner, Georg</au><au>Bischof, Horst</au><au>Hornik, Kurt</au><au>Hornik, Kurt</au><au>Dorffner, Georg</au><au>Bischof, Horst</au><format>book</format><genre>bookitem</genre><ristype>CHAP</ristype><atitle>Online Symbolic-Sequence Prediction with Discrete-Time Recurrent Neural Networks</atitle><btitle>Artificial Neural Networks - ICANN 2001</btitle><seriestitle>Lecture Notes in Computer Science</seriestitle><date>2001</date><risdate>2001</risdate><volume>2130</volume><spage>719</spage><epage>724</epage><pages>719-724</pages><issn>0302-9743</issn><eissn>1611-3349</eissn><isbn>3540424865</isbn><isbn>9783540424864</isbn><eisbn>3540446680</eisbn><eisbn>9783540446682</eisbn><abstract>This paper studies the use of discrete-time recurrent neural networks for predicting the next symbol in a sequence. The focus is on online prediction, a task much harder than the classical o.ine grammatical inference with neural networks. The results obtained show that the performance of recurrent networks working online is acceptable when sequences come from finite-state machines or even from some chaotic sources. When predicting texts in human language, however, dynamics seem to be too complex to be correctly learned in real-time by the net. Two algorithms are considered for network training: real-time recurrent learning and the decoupled extended Kalman filter.</abstract><cop>Germany</cop><pub>Springer Berlin / Heidelberg</pub><doi>10.1007/3-540-44668-0_100</doi><oclcid>958524338</oclcid><tpages>6</tpages><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 0302-9743
ispartof Artificial Neural Networks - ICANN 2001, 2001, Vol.2130, p.719-724
issn 0302-9743
1611-3349
language eng
recordid cdi_pascalfrancis_primary_14045253
source Springer Books
subjects Applied sciences
Artificial intelligence
Computer science
control theory
systems
Connectionism. Neural networks
Exact sciences and technology
title Online Symbolic-Sequence Prediction with Discrete-Time Recurrent Neural Networks
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-30T18%3A15%3A06IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pasca&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=bookitem&rft.atitle=Online%20Symbolic-Sequence%20Prediction%20with%20Discrete-Time%20Recurrent%20Neural%20Networks&rft.btitle=Artificial%20Neural%20Networks%20-%20ICANN%202001&rft.au=P%C3%A9rez-Ortiz,%20Juan%20Antonio&rft.date=2001&rft.volume=2130&rft.spage=719&rft.epage=724&rft.pages=719-724&rft.issn=0302-9743&rft.eissn=1611-3349&rft.isbn=3540424865&rft.isbn_list=9783540424864&rft_id=info:doi/10.1007/3-540-44668-0_100&rft_dat=%3Cproquest_pasca%3EEBC3073004_106_721%3C/proquest_pasca%3E%3Curl%3E%3C/url%3E&rft.eisbn=3540446680&rft.eisbn_list=9783540446682&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=EBC3073004_106_721&rft_id=info:pmid/&rfr_iscdi=true