Stable learning in stochastic network states

The mammalian cerebral cortex is characterized in vivo by irregular spontaneous activity, but how this ongoing dynamics affects signal processing and learning remains unknown. The associative plasticity rules demonstrated in vitro, mostly in silent networks, are based on the detection of correlation...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:The Journal of neuroscience 2012-01, Vol.32 (1), p.194-214
Hauptverfasser: El Boustani, Sami, Yger, Pierre, Frégnac, Yves, Destexhe, Alain
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 214
container_issue 1
container_start_page 194
container_title The Journal of neuroscience
container_volume 32
creator El Boustani, Sami
Yger, Pierre
Frégnac, Yves
Destexhe, Alain
description The mammalian cerebral cortex is characterized in vivo by irregular spontaneous activity, but how this ongoing dynamics affects signal processing and learning remains unknown. The associative plasticity rules demonstrated in vitro, mostly in silent networks, are based on the detection of correlations between presynaptic and postsynaptic activity and hence are sensitive to spontaneous activity and spurious correlations. Therefore, they cannot operate in realistic network states. Here, we present a new class of spike-timing-dependent plasticity learning rules with local floating plasticity thresholds, the slow dynamics of which account for metaplasticity. This novel algorithm is shown to both correctly predict homeostasis in synaptic weights and solve the problem of asymptotic stable learning in noisy states. It is shown to naturally encompass many other known types of learning rule, unifying them into a single coherent framework. The mixed presynaptic and postsynaptic dependency of the floating plasticity threshold is justified by a cascade of known molecular pathways, which leads to experimentally testable predictions.
doi_str_mv 10.1523/jneurosci.2496-11.2012
format Article
fullrecord <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_6621309</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>914663657</sourcerecordid><originalsourceid>FETCH-LOGICAL-c513t-7c170012ec034d4ccf75a1710bba04373e9b836d0ee5efa2127fc18f8e59b5033</originalsourceid><addsrcrecordid>eNpdUdtKw0AQXUSx9fILkjcRTJ3Z3ewmL0IpVStFwcvzstlObGqa1Gyq-PcmtIr6NHDmnDOXw9gJwgAjLi4WJa3ryrt8wGWiQsQBB-Q7rN92k5BLwF3WB64hVFLLHjvwfgEAGlDvsx7nHBMe8z47f2xsWlBQkK3LvHwJ8jLwTeXm1je5C0pqPqr6tYVsQ_6I7WW28HS8rYfs-Wr8NLoJp_fXk9FwGroIRRNqhxrabciBkDPpXKYjixohTS1IoQUlaSzUDIgiyixHrjOHcRZTlKQRCHHILje-q3W6pJmjsqltYVZ1vrT1p6lsbv52ynxuXqp3oxRHAUlrcLYxmP-T3QynpsMAVCwhlu_Yck-3w-rqbU2-McvcOyoKW1K19iZBqZRQkW6ZasN07ed9TdmPNYLpUjG3d-Pnh_vH0cR0qRhE06XSCk9-3_Mj-45BfAF1MIn_</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>914663657</pqid></control><display><type>article</type><title>Stable learning in stochastic network states</title><source>MEDLINE</source><source>Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals</source><source>PubMed Central</source><creator>El Boustani, Sami ; Yger, Pierre ; Frégnac, Yves ; Destexhe, Alain</creator><creatorcontrib>El Boustani, Sami ; Yger, Pierre ; Frégnac, Yves ; Destexhe, Alain</creatorcontrib><description>The mammalian cerebral cortex is characterized in vivo by irregular spontaneous activity, but how this ongoing dynamics affects signal processing and learning remains unknown. The associative plasticity rules demonstrated in vitro, mostly in silent networks, are based on the detection of correlations between presynaptic and postsynaptic activity and hence are sensitive to spontaneous activity and spurious correlations. Therefore, they cannot operate in realistic network states. Here, we present a new class of spike-timing-dependent plasticity learning rules with local floating plasticity thresholds, the slow dynamics of which account for metaplasticity. This novel algorithm is shown to both correctly predict homeostasis in synaptic weights and solve the problem of asymptotic stable learning in noisy states. It is shown to naturally encompass many other known types of learning rule, unifying them into a single coherent framework. The mixed presynaptic and postsynaptic dependency of the floating plasticity threshold is justified by a cascade of known molecular pathways, which leads to experimentally testable predictions.</description><identifier>ISSN: 0270-6474</identifier><identifier>EISSN: 1529-2401</identifier><identifier>DOI: 10.1523/jneurosci.2496-11.2012</identifier><identifier>PMID: 22219282</identifier><language>eng</language><publisher>United States: Society for Neuroscience</publisher><subject>Action Potentials ; Action Potentials - physiology ; Algorithms ; Animals ; Cerebral Cortex ; Cerebral Cortex - physiology ; Humans ; Learning ; Learning - physiology ; Life Sciences ; Models, Neurological ; Nerve Net ; Nerve Net - physiology ; Neural Networks (Computer) ; Neuronal Plasticity ; Neuronal Plasticity - physiology ; Neurons ; Neurons - physiology ; Neurons and Cognition ; Stochastic Processes</subject><ispartof>The Journal of neuroscience, 2012-01, Vol.32 (1), p.194-214</ispartof><rights>Distributed under a Creative Commons Attribution 4.0 International License</rights><rights>Copyright © 2012 the authors 0270-6474/12/320194-21$15.00/0 2012</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c513t-7c170012ec034d4ccf75a1710bba04373e9b836d0ee5efa2127fc18f8e59b5033</citedby><cites>FETCH-LOGICAL-c513t-7c170012ec034d4ccf75a1710bba04373e9b836d0ee5efa2127fc18f8e59b5033</cites><orcidid>0000-0001-7405-0455</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC6621309/pdf/$$EPDF$$P50$$Gpubmedcentral$$H</linktopdf><linktohtml>$$Uhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC6621309/$$EHTML$$P50$$Gpubmedcentral$$H</linktohtml><link.rule.ids>230,315,729,782,786,887,27931,27932,53798,53800</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/22219282$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink><backlink>$$Uhttps://hal.science/hal-00684084$$DView record in HAL$$Hfree_for_read</backlink></links><search><creatorcontrib>El Boustani, Sami</creatorcontrib><creatorcontrib>Yger, Pierre</creatorcontrib><creatorcontrib>Frégnac, Yves</creatorcontrib><creatorcontrib>Destexhe, Alain</creatorcontrib><title>Stable learning in stochastic network states</title><title>The Journal of neuroscience</title><addtitle>J Neurosci</addtitle><description>The mammalian cerebral cortex is characterized in vivo by irregular spontaneous activity, but how this ongoing dynamics affects signal processing and learning remains unknown. The associative plasticity rules demonstrated in vitro, mostly in silent networks, are based on the detection of correlations between presynaptic and postsynaptic activity and hence are sensitive to spontaneous activity and spurious correlations. Therefore, they cannot operate in realistic network states. Here, we present a new class of spike-timing-dependent plasticity learning rules with local floating plasticity thresholds, the slow dynamics of which account for metaplasticity. This novel algorithm is shown to both correctly predict homeostasis in synaptic weights and solve the problem of asymptotic stable learning in noisy states. It is shown to naturally encompass many other known types of learning rule, unifying them into a single coherent framework. The mixed presynaptic and postsynaptic dependency of the floating plasticity threshold is justified by a cascade of known molecular pathways, which leads to experimentally testable predictions.</description><subject>Action Potentials</subject><subject>Action Potentials - physiology</subject><subject>Algorithms</subject><subject>Animals</subject><subject>Cerebral Cortex</subject><subject>Cerebral Cortex - physiology</subject><subject>Humans</subject><subject>Learning</subject><subject>Learning - physiology</subject><subject>Life Sciences</subject><subject>Models, Neurological</subject><subject>Nerve Net</subject><subject>Nerve Net - physiology</subject><subject>Neural Networks (Computer)</subject><subject>Neuronal Plasticity</subject><subject>Neuronal Plasticity - physiology</subject><subject>Neurons</subject><subject>Neurons - physiology</subject><subject>Neurons and Cognition</subject><subject>Stochastic Processes</subject><issn>0270-6474</issn><issn>1529-2401</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2012</creationdate><recordtype>article</recordtype><sourceid>EIF</sourceid><recordid>eNpdUdtKw0AQXUSx9fILkjcRTJ3Z3ewmL0IpVStFwcvzstlObGqa1Gyq-PcmtIr6NHDmnDOXw9gJwgAjLi4WJa3ryrt8wGWiQsQBB-Q7rN92k5BLwF3WB64hVFLLHjvwfgEAGlDvsx7nHBMe8z47f2xsWlBQkK3LvHwJ8jLwTeXm1je5C0pqPqr6tYVsQ_6I7WW28HS8rYfs-Wr8NLoJp_fXk9FwGroIRRNqhxrabciBkDPpXKYjixohTS1IoQUlaSzUDIgiyixHrjOHcRZTlKQRCHHILje-q3W6pJmjsqltYVZ1vrT1p6lsbv52ynxuXqp3oxRHAUlrcLYxmP-T3QynpsMAVCwhlu_Yck-3w-rqbU2-McvcOyoKW1K19iZBqZRQkW6ZasN07ed9TdmPNYLpUjG3d-Pnh_vH0cR0qRhE06XSCk9-3_Mj-45BfAF1MIn_</recordid><startdate>20120104</startdate><enddate>20120104</enddate><creator>El Boustani, Sami</creator><creator>Yger, Pierre</creator><creator>Frégnac, Yves</creator><creator>Destexhe, Alain</creator><general>Society for Neuroscience</general><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><scope>1XC</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0001-7405-0455</orcidid></search><sort><creationdate>20120104</creationdate><title>Stable learning in stochastic network states</title><author>El Boustani, Sami ; Yger, Pierre ; Frégnac, Yves ; Destexhe, Alain</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c513t-7c170012ec034d4ccf75a1710bba04373e9b836d0ee5efa2127fc18f8e59b5033</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2012</creationdate><topic>Action Potentials</topic><topic>Action Potentials - physiology</topic><topic>Algorithms</topic><topic>Animals</topic><topic>Cerebral Cortex</topic><topic>Cerebral Cortex - physiology</topic><topic>Humans</topic><topic>Learning</topic><topic>Learning - physiology</topic><topic>Life Sciences</topic><topic>Models, Neurological</topic><topic>Nerve Net</topic><topic>Nerve Net - physiology</topic><topic>Neural Networks (Computer)</topic><topic>Neuronal Plasticity</topic><topic>Neuronal Plasticity - physiology</topic><topic>Neurons</topic><topic>Neurons - physiology</topic><topic>Neurons and Cognition</topic><topic>Stochastic Processes</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>El Boustani, Sami</creatorcontrib><creatorcontrib>Yger, Pierre</creatorcontrib><creatorcontrib>Frégnac, Yves</creatorcontrib><creatorcontrib>Destexhe, Alain</creatorcontrib><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><collection>Hyper Article en Ligne (HAL)</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>The Journal of neuroscience</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>El Boustani, Sami</au><au>Yger, Pierre</au><au>Frégnac, Yves</au><au>Destexhe, Alain</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Stable learning in stochastic network states</atitle><jtitle>The Journal of neuroscience</jtitle><addtitle>J Neurosci</addtitle><date>2012-01-04</date><risdate>2012</risdate><volume>32</volume><issue>1</issue><spage>194</spage><epage>214</epage><pages>194-214</pages><issn>0270-6474</issn><eissn>1529-2401</eissn><abstract>The mammalian cerebral cortex is characterized in vivo by irregular spontaneous activity, but how this ongoing dynamics affects signal processing and learning remains unknown. The associative plasticity rules demonstrated in vitro, mostly in silent networks, are based on the detection of correlations between presynaptic and postsynaptic activity and hence are sensitive to spontaneous activity and spurious correlations. Therefore, they cannot operate in realistic network states. Here, we present a new class of spike-timing-dependent plasticity learning rules with local floating plasticity thresholds, the slow dynamics of which account for metaplasticity. This novel algorithm is shown to both correctly predict homeostasis in synaptic weights and solve the problem of asymptotic stable learning in noisy states. It is shown to naturally encompass many other known types of learning rule, unifying them into a single coherent framework. The mixed presynaptic and postsynaptic dependency of the floating plasticity threshold is justified by a cascade of known molecular pathways, which leads to experimentally testable predictions.</abstract><cop>United States</cop><pub>Society for Neuroscience</pub><pmid>22219282</pmid><doi>10.1523/jneurosci.2496-11.2012</doi><tpages>21</tpages><orcidid>https://orcid.org/0000-0001-7405-0455</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 0270-6474
ispartof The Journal of neuroscience, 2012-01, Vol.32 (1), p.194-214
issn 0270-6474
1529-2401
language eng
recordid cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_6621309
source MEDLINE; Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals; PubMed Central
subjects Action Potentials
Action Potentials - physiology
Algorithms
Animals
Cerebral Cortex
Cerebral Cortex - physiology
Humans
Learning
Learning - physiology
Life Sciences
Models, Neurological
Nerve Net
Nerve Net - physiology
Neural Networks (Computer)
Neuronal Plasticity
Neuronal Plasticity - physiology
Neurons
Neurons - physiology
Neurons and Cognition
Stochastic Processes
title Stable learning in stochastic network states
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-04T05%3A37%3A42IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Stable%20learning%20in%20stochastic%20network%20states&rft.jtitle=The%20Journal%20of%20neuroscience&rft.au=El%20Boustani,%20Sami&rft.date=2012-01-04&rft.volume=32&rft.issue=1&rft.spage=194&rft.epage=214&rft.pages=194-214&rft.issn=0270-6474&rft.eissn=1529-2401&rft_id=info:doi/10.1523/jneurosci.2496-11.2012&rft_dat=%3Cproquest_pubme%3E914663657%3C/proquest_pubme%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=914663657&rft_id=info:pmid/22219282&rfr_iscdi=true