EqSpike: Spike-driven equilibrium propagation for neuromorphic implementations

Finding spike-based learning algorithms that can be implemented within the local constraints of neuromorphic systems, while achieving high accuracy, remains a formidable challenge. Equilibrium propagation is a promising alternative to backpropagation as it only involves local computations, but hardw...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	iScience Jg. 24; H. 3; S. 102222
Hauptverfasser:	Martin, Erwann, Ernoult, Maxence, Laydevant, Jérémie, Li, Shuai, Querlioz, Damien, Petrisor, Teodora, Grollier, Julie
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	United States Elsevier Inc 19.03.2021 Elsevier
Schlagworte:	Algorithms Artificial Intelligence Computer Science Physics Artificial Intelligence Algorithms Computer Science
ISSN:	2589-0042, 2589-0042
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Abstract	Finding spike-based learning algorithms that can be implemented within the local constraints of neuromorphic systems, while achieving high accuracy, remains a formidable challenge. Equilibrium propagation is a promising alternative to backpropagation as it only involves local computations, but hardware-oriented studies have so far focused on rate-based networks. In this work, we develop a spiking neural network algorithm called EqSpike, compatible with neuromorphic systems, which learns by equilibrium propagation. Through simulations, we obtain a test recognition accuracy of 97.6% on the MNIST handwritten digits dataset (Mixed National Institute of Standards and Technology), similar to rate-based equilibrium propagation, and comparing favorably to alternative learning techniques for spiking neural networks. We show that EqSpike implemented in silicon neuromorphic technology could reduce the energy consumption of inference and training, respectively, by three orders and two orders of magnitude compared to graphics processing units. Finally, we also show that during learning, EqSpike weight updates exhibit a form of spike-timing-dependent plasticity, highlighting a possible connection with biology. [Display omitted] •EqSpike is a spiking neural network version of equilibrium propagation•It achieves 97.6% test accuracy on MNIST with a fully connected architecture•Its two-factor local learning rule is compatible with neuromorphic hardware•Its weight updates exhibit a form of spike-timing-dependent plasticity Computer Science; Algorithms; Artificial Intelligence
AbstractList	Finding spike-based learning algorithms that can be implemented within the local constraints of neuromorphic systems, while achieving high accuracy, remains a formidable challenge. Equilibrium propagation is a promising alternative to backpropagation as it only involves local computations, but hardware-oriented studies have so far focused on rate-based networks. In this work, we develop a spiking neural network algorithm called EqSpike, compatible with neuromorphic systems, which learns by equilibrium propagation. Through simulations, we obtain a test recognition accuracy of 97.6% on the MNIST handwritten digits dataset (Mixed National Institute of Standards and Technology), similar to rate-based equilibrium propagation, and comparing favorably to alternative learning techniques for spiking neural networks. We show that EqSpike implemented in silicon neuromorphic technology could reduce the energy consumption of inference and training, respectively, by three orders and two orders of magnitude compared to graphics processing units. Finally, we also show that during learning, EqSpike weight updates exhibit a form of spike-timing-dependent plasticity, highlighting a possible connection with biology. Finding spike-based learning algorithms that can be implemented within the local constraints of neuromorphic systems, while achieving high accuracy, remains a formidable challenge. Equilibrium propagation is a promising alternative to backpropagation as it only involves local computations, but hardware-oriented studies have so far focused on rate-based networks. In this work, we develop a spiking neural network algorithm called EqSpike, compatible with neuromorphic systems, which learns by equilibrium propagation. Through simulations, we obtain a test recognition accuracy of 97.6% on the MNIST handwritten digits dataset (Mixed National Institute of Standards and Technology), similar to rate-based equilibrium propagation, and comparing favorably to alternative learning techniques for spiking neural networks. We show that EqSpike implemented in silicon neuromorphic technology could reduce the energy consumption of inference and training, respectively, by three orders and two orders of magnitude compared to graphics processing units. Finally, we also show that during learning, EqSpike weight updates exhibit a form of spike-timing-dependent plasticity, highlighting a possible connection with biology. • EqSpike is a spiking neural network version of equilibrium propagation • It achieves 97.6% test accuracy on MNIST with a fully connected architecture • Its two-factor local learning rule is compatible with neuromorphic hardware • Its weight updates exhibit a form of spike-timing-dependent plasticity Computer Science; Algorithms; Artificial Intelligence Finding spike-based learning algorithms that can be implemented within the local constraints of neuromorphic systems, while achieving high accuracy, remains a formidable challenge. Equilibrium propagation is a promising alternative to backpropagation as it only involves local computations, but hardware-oriented studies have so far focused on rate-based networks. In this work, we develop a spiking neural network algorithm called EqSpike, compatible with neuromorphic systems, which learns by equilibrium propagation. Through simulations, we obtain a test recognition accuracy of 97.6% on the MNIST handwritten digits dataset (Mixed National Institute of Standards and Technology), similar to rate-based equilibrium propagation, and comparing favorably to alternative learning techniques for spiking neural networks. We show that EqSpike implemented in silicon neuromorphic technology could reduce the energy consumption of inference and training, respectively, by three orders and two orders of magnitude compared to graphics processing units. Finally, we also show that during learning, EqSpike weight updates exhibit a form of spike-timing-dependent plasticity, highlighting a possible connection with biology.Finding spike-based learning algorithms that can be implemented within the local constraints of neuromorphic systems, while achieving high accuracy, remains a formidable challenge. Equilibrium propagation is a promising alternative to backpropagation as it only involves local computations, but hardware-oriented studies have so far focused on rate-based networks. In this work, we develop a spiking neural network algorithm called EqSpike, compatible with neuromorphic systems, which learns by equilibrium propagation. Through simulations, we obtain a test recognition accuracy of 97.6% on the MNIST handwritten digits dataset (Mixed National Institute of Standards and Technology), similar to rate-based equilibrium propagation, and comparing favorably to alternative learning techniques for spiking neural networks. We show that EqSpike implemented in silicon neuromorphic technology could reduce the energy consumption of inference and training, respectively, by three orders and two orders of magnitude compared to graphics processing units. Finally, we also show that during learning, EqSpike weight updates exhibit a form of spike-timing-dependent plasticity, highlighting a possible connection with biology. Finding spike-based learning algorithms that can be implemented within the local constraints of neuromorphic systems, while achieving high accuracy, remains a formidable challenge. Equilibrium propagation is a promising alternative to backpropagation as it only involves local computations, but hardware-oriented studies have so far focused on rate-based networks. In this work, we develop a spiking neural network algorithm called EqSpike, compatible with neuromorphic systems, which learns by equilibrium propagation. Through simulations, we obtain a test recognition accuracy of 97.6% on the MNIST handwritten digits dataset (Mixed National Institute of Standards and Technology), similar to rate-based equilibrium propagation, and comparing favorably to alternative learning techniques for spiking neural networks. We show that EqSpike implemented in silicon neuromorphic technology could reduce the energy consumption of inference and training, respectively, by three orders and two orders of magnitude compared to graphics processing units. Finally, we also show that during learning, EqSpike weight updates exhibit a form of spike-timing-dependent plasticity, highlighting a possible connection with biology. [Display omitted] •EqSpike is a spiking neural network version of equilibrium propagation•It achieves 97.6% test accuracy on MNIST with a fully connected architecture•Its two-factor local learning rule is compatible with neuromorphic hardware•Its weight updates exhibit a form of spike-timing-dependent plasticity Computer Science; Algorithms; Artificial Intelligence
ArticleNumber	102222
Author	Martin, Erwann Grollier, Julie Ernoult, Maxence Querlioz, Damien Petrisor, Teodora Li, Shuai Laydevant, Jérémie
Author_xml	– sequence: 1 givenname: Erwann surname: Martin fullname: Martin, Erwann organization: Thales Research and Technology, 91767 Palaiseau, France – sequence: 2 givenname: Maxence surname: Ernoult fullname: Ernoult, Maxence organization: Unité Mixte de Physique, CNRS, Thales, Université Paris-Saclay, 91767 Palaiseau, France – sequence: 3 givenname: Jérémie surname: Laydevant fullname: Laydevant, Jérémie organization: Unité Mixte de Physique, CNRS, Thales, Université Paris-Saclay, 91767 Palaiseau, France – sequence: 4 givenname: Shuai surname: Li fullname: Li, Shuai organization: Unité Mixte de Physique, CNRS, Thales, Université Paris-Saclay, 91767 Palaiseau, France – sequence: 5 givenname: Damien surname: Querlioz fullname: Querlioz, Damien organization: Université Paris-Saclay, CNRS, Centre de Nanosciences et de Nanotechnologies, 91120 Palaiseau, France – sequence: 6 givenname: Teodora surname: Petrisor fullname: Petrisor, Teodora organization: Thales Research and Technology, 91767 Palaiseau, France – sequence: 7 givenname: Julie surname: Grollier fullname: Grollier, Julie email: julie.grollier@cnrs-thales.fr organization: Unité Mixte de Physique, CNRS, Thales, Université Paris-Saclay, 91767 Palaiseau, France
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/33748709$$D View this record in MEDLINE/PubMed https://hal.science/hal-03451693$$DView record in HAL
BookMark	eNp9kk1v1DAQhi1URNulf4ADyhEOWfwZOwghVVVLK63gAJwtrz3Z9ZLEWSdZiX-Ps2lR20MtS2PNzPt47JlzdNKGFhB6R_CSYFJ82i19b_2SYkqSg6b1Cp1RococY05PHp1P0UXf7zDGNG1eFm_QKWOSK4nLM_T9ev-z83_gc3Y0uYv-AG0G-9HXfh392GRdDJ3ZmMGHNqtCzFoYY2hC7LbeZr7pamigHY7x_i16XZm6h4t7u0C_b65_Xd3mqx_f7q4uV7kVQgw5A75mklJrKseUFEJxVZXYWllYQUusnHCOYwGWK1ZxI40ljljOlBWYYMcW6G7mumB2uou-MfGvDsbroyPEjTZx8LYGTUQFtiKKyYLxwhUTYm1LuTZKYWYn1teZ1Y3rBpxNj4mmfgJ9Gmn9Vm_CQctSYlaQBPg4A7bPZLeXKz35MOOCFCU7TLkf7i-LYT9CP-gm9RHq2rQQxl5TkZCK4mQW6P3juv6TH5qXEtScYGPo-wiVtn7uQyrT15pgPY2K3ulpVPQ0KnoelSSlz6QP9BdFX2YRpM4ePESdMqC14HwEO6Sv9y_J_wFGH9cs
CitedBy_id	crossref_primary_10_1038_s43588_021_00184_y crossref_primary_10_1103_PhysRevApplied_17_034077 crossref_primary_10_3390_nano15030213 crossref_primary_10_1109_TCAD_2023_3285410 crossref_primary_10_3390_math13111866 crossref_primary_10_1146_annurev_conmatphys_040821_113439 crossref_primary_10_3389_fnins_2022_736642 crossref_primary_10_1038_s41467_023_43887_8 crossref_primary_10_1038_s41586_021_04223_6 crossref_primary_10_1103_PhysRevApplied_18_014040 crossref_primary_10_1016_j_neucom_2022_10_068 crossref_primary_10_1038_s41467_025_57043_x crossref_primary_10_1103_PhysRevX_13_031020 crossref_primary_10_1103_smt9_1t1l crossref_primary_10_3390_mi14071367 crossref_primary_10_1103_PhysRevResearch_4_L022037 crossref_primary_10_1038_s41467_025_61665_6 crossref_primary_10_1088_2399_1984_ad299a crossref_primary_10_3389_fncom_2023_1114651 crossref_primary_10_1038_s41567_024_02534_9 crossref_primary_10_1038_s41467_024_46879_4 crossref_primary_10_1088_2634_4386_acdf17 crossref_primary_10_1088_2634_4386_ad752b crossref_primary_10_1088_2634_4386_ad8c78 crossref_primary_10_1002_adma_202203352 crossref_primary_10_1007_s40509_024_00351_6 crossref_primary_10_1088_2634_4386_ad3a95 crossref_primary_10_1103_PhysRevApplied_22_024053 crossref_primary_10_1109_JPROC_2023_3273520 crossref_primary_10_1016_j_neunet_2023_01_026
Cites_doi	10.3389/fnins.2015.00141 10.3389/fnins.2021.633674 10.3389/fnins.2020.00424 10.3389/fnins.2017.00324 10.1016/j.patcog.2019.04.016 10.1038/s41578-019-0159-3 10.1038/s41928-020-0435-7 10.1038/ncomms12611 10.1162/NECO_a_00934 10.3389/fnins.2018.00774 10.1021/nl904092h 10.1016/j.neunet.2012.02.022 10.1038/s41586-019-1157-8 10.1142/S0129065720500276 10.3389/fnins.2011.00026 10.1038/s41467-018-07757-y 10.7554/eLife.54940 10.3389/fnins.2020.00240 10.1038/s41598-017-05480-0 10.1109/JPROC.2014.2304638 10.1146/annurev.neuro.24.1.139 10.1126/science.1254642 10.3389/fnins.2018.00991 10.3390/ma13010166 10.3389/fncom.2017.00024 10.1038/381526a0 10.1016/j.neucom.2018.11.014 10.1109/MSP.2019.2931595 10.1038/s41593-019-0520-2 10.1109/JSSC.2019.2942367 10.3389/fnins.2016.00508 10.1109/MM.2018.112130359 10.1063/1.4921745 10.1371/journal.pcbi.0030031 10.1038/s41467-020-17236-y
ContentType	Journal Article
Copyright	2021 The Authors 2021 The Authors. Attribution - NonCommercial 2021 The Authors 2021
Copyright_xml	– notice: 2021 The Authors – notice: 2021 The Authors. – notice: Attribution - NonCommercial – notice: 2021 The Authors 2021
DBID	6I. AAFTH AAYXX CITATION NPM 7X8 1XC VOOES 5PM DOA
DOI	10.1016/j.isci.2021.102222
DatabaseName	ScienceDirect Open Access Titles Elsevier:ScienceDirect:Open Access CrossRef PubMed MEDLINE - Academic Hyper Article en Ligne (HAL) Hyper Article en Ligne (HAL) (Open Access) PubMed Central (Full Participant titles) DOAJ Directory of Open Access Journals
DatabaseTitle	CrossRef PubMed MEDLINE - Academic
DatabaseTitleList	MEDLINE - Academic PubMed
Database_xml	– sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website – sequence: 2 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 3 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database
DeliveryMethod	fulltext_linktorsrc
Discipline	Computer Science Physics
EISSN	2589-0042
ExternalDocumentID	oai_doaj_org_article_15fecf18376346d68c50bc97ba8803cd PMC7970361 oai:HAL:hal-03451693v1 33748709 10_1016_j_isci_2021_102222 S2589004221001905
Genre	Journal Article
GroupedDBID	0R~ 53G 6I. AACTN AAEDW AAFTH AALRI AAMRU AAXUO ABMAC ADBBV ADVLN AEXQZ AFTJW AITUG ALMA_UNASSIGNED_HOLDINGS AMRAJ AOIJS BCNDV EBS FDB GROUPED_DOAJ HYE M41 NCXOZ OK1 ROL RPM SSZ AAYWO AAYXX ACVFH ADCNI AEUPX AFPUW AIGII AKBMS AKYEP APXCP CITATION EJD NPM 7X8 1XC VOOES 5PM
ID	FETCH-LOGICAL-c555t-3e4b3722cafd38755848f90cc76c52908d5dd405ec483f4a7ac1d1c438c5010d3
IEDL.DBID	DOA
ISICitedReferencesCount	42
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000631646000072&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	2589-0042
IngestDate	Fri Oct 03 12:51:26 EDT 2025 Tue Sep 30 16:50:37 EDT 2025 Tue Oct 14 20:23:05 EDT 2025 Fri Jul 11 08:52:49 EDT 2025 Thu Jan 02 22:36:57 EST 2025 Tue Nov 18 22:45:15 EST 2025 Sat Nov 29 02:13:31 EST 2025 Sat Nov 16 15:58:52 EST 2024
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	3
Keywords	Artificial Intelligence Algorithms Computer Science
Language	English
License	This is an open access article under the CC BY license. 2021 The Authors. Attribution - NonCommercial: http://creativecommons.org/licenses/by-nc This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c555t-3e4b3722cafd38755848f90cc76c52908d5dd405ec483f4a7ac1d1c438c5010d3
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 PMCID: PMC7970361
ORCID	0000-0003-4866-4490
OpenAccessLink	https://doaj.org/article/15fecf18376346d68c50bc97ba8803cd
PMID	33748709
PQID	2503682050
PQPubID	23479
ParticipantIDs	doaj_primary_oai_doaj_org_article_15fecf18376346d68c50bc97ba8803cd pubmedcentral_primary_oai_pubmedcentral_nih_gov_7970361 hal_primary_oai_HAL_hal_03451693v1 proquest_miscellaneous_2503682050 pubmed_primary_33748709 crossref_citationtrail_10_1016_j_isci_2021_102222 crossref_primary_10_1016_j_isci_2021_102222 elsevier_sciencedirect_doi_10_1016_j_isci_2021_102222
PublicationCentury	2000
PublicationDate	2021-03-19
PublicationDateYYYYMMDD	2021-03-19
PublicationDate_xml	– month: 03 year: 2021 text: 2021-03-19 day: 19
PublicationDecade	2020
PublicationPlace	United States
PublicationPlace_xml	– name: United States
PublicationTitle	iScience
PublicationTitleAlternate	iScience
PublicationYear	2021
Publisher	Elsevier Inc Elsevier
Publisher_xml	– name: Elsevier Inc – name: Elsevier
References	Tavanaei, Maida (bib54) 2019; 330 Milo, Malavena, Monzio Compagnoni, Ielmini (bib32) 2020; 13 Thiele, Bichler, Dupret (bib56) 2019 Ishii, Kim, Lewis, Okazaki, Okazawa, Ito, Rasch, Kim, Nomura, Shin (bib16) 2019 Ernoult, Grollier, Querlioz, Bengio, Scellier (bib6) 2019 Ernoult, Grollier, Querlioz, Bengio, Scellier (bib7) 2020 Zoppo, Marrone, Corinto (bib61) 2020; 14 Schemmel, Brüderle, Grübl, Hock, Meier, Millner (bib51) 2010 Laborieux, Ernoult, Scellier, Bengio, Grollier, Querlioz (bib23) 2021; 15 Qiao, Mostafa, Corradi, Osswald, Stefanini, Sumislawska, Indiveri (bib45) 2015; 9 Zhang, Gao, Tang, Yao, Yu, Chang, Yoo, Qian, Wu (bib60) 2020; 3 Bellec, Scherr, Subramoney, Hajek, Salaj, Legenstein, Maass (bib1) 2020; 11 Neftci, Mostafa, Zenke (bib36) 2019; 36 Kheradpisheh, Masquelier (bib21) 2020; 30 Prezioso, Mahmoodi, Merrikh Bayat, Nili, Kim, Vincent, Strukov (bib44) 2018; 9 Wang, Wu, Burr, Hwang, Wang, Xia, Yang (bib58) 2020; 5 Pfeiffer, Pfeil (bib43) 2018; 12 Bengio, Mesnard, Fischer, Zhang, Wu (bib2) 2017; 29 Merolla, Arthur, Alvarez-Icaza, Cassidy, Sawada, Akopyan, Jackson, Imam, Guo, Nakamura (bib30) 2014; 345 Kaiser, Mostafa, Neftci (bib19) 2020; 14 Richards, Lillicrap, Beaudoin, Bengio, Bogacz, Christensen, Clopath, Ponte Costa, de Berker, Ganguli (bib47) 2019; 22 Huh, Sejnowski (bib15) 2018 Sebastian, Le Gallo, Khaddam-Aljameh, Eleftheriou (bib52) 2020; 15 Davies, Srinivasa, Lin, Chinya, Cao, Choday, Dimou, Joshi, Imam, Jain (bib5) 2018; 38 Gerstner (bib12) 2014 Sacramento, Ponte Costa, Bengio, Senn (bib48) 2018; 31 Hirtzlin, Bocquet, Ernoult, Klein, Nowak, Vianello, Portal, Querlioz (bib13) 2019 Kendall, Pantone, Manickavasagam, Bengio, Scellier (bib20) 2020 (bib29) 1989 O’Connor, Welling (bib38) 2016 Thakur, Lottier Molin, Cauwenberghs, Indiveri, Kumar, Qiao, Schemmel, Wang, Chicca, Hasler (bib55) 2018; 12 Zamarreño-Ramos, Camuñas-Mesa, Pérez-Carrasco, Masquelier, Serrano-Gotarredona, Linares-Barranco (bib26) 2011; 5 Bi, Poo (bib3) 2001; 24 Joseph, V. and Nagarajan, C. (2020 )MADONNA: A Framework for Energy Measurements and Assistance in Designing Low Power Deep Neural Networks. p. 7. Li, Liu, Nandi, Venkatachalam, Elliman (bib25) 2015; 106 Furber, Galluppi, Temple, Plana (bib11) 2014; 102 Neftci, Augustine, Paul, Detorakis (bib35) 2017; 11 Rastegari, Ordonez, Redmon, Farhadi (bib46) 2016 Scellier, Bengio (bib50) 2017; 11 Masquelier, Thorpe (bib28) 2007; 3 Mostafa (bib33) 2018; 29 Hubara, Courbariaux, Soudry, El-Yaniv, Bengio (bib14) 2016 Navarro, Salari, Lin, Cowan, Penington, Milescu, Milescu (bib34) 2020; 9 Park, Lee, Jeon (bib39) 2020; 55 Payeur, Guerguiev, Zenke, Richards, Naud (bib40) 2020 Lee, Delbruck, Pfeiffer (bib24) 2016; 10 Xi, Gao, Tang, Chen, Chang, Hu, Van Der Spiegel, Qian, Wu (bib59) 2020; 109 Feldmann, Youngblood, Wright, Bhaskaran, Pernice (bib9) 2019; 569 Marković, Mizrahi, Querlioz, Grollier (bib27) 2020 Jo, Chang, Ebong, Bhadviya, Mazumder, Lu (bib17) 2010; 10 Scellier, Goyal, Binas, Mesnard, Bengio (bib49) 2018 Wan, Kubendran, Burc Eryilmaz, Zhang, Liao, Wu, Deiss, Gao, Raina, Joshi (bib57) 2020 Payvand, Fouda, Kurdahi, Eltawil, Neftci (bib41) 2020 O’Connor, Gavves, Welling (bib37) 2019 Falez, Tirilly, Bilasco, Devienne, Boulet (bib8) 2019; 93 Kirkwood, Rioult, Bear (bib22) 1996; 381 Mesnard, Gerstner, Brea (bib31) 2016 Frenkel, Lefebvre, Legat, Bol (bib10) 2019; 13 Pedretti, Milo, Ambrogio, Carboni, Bianchi, Calderoni, Ramaswamy, Spinelli, Ielmini (bib42) 2017; 7 Bichler, Querlioz, Thorpe, Bourgoin, Gamrat (bib4) 2012; 32 Serb, Bill, Khiat, Berdan, Legenstein, Prodromakis (bib53) 2016; 7 Ernoult (10.1016/j.isci.2021.102222_bib7) 2020 Pfeiffer (10.1016/j.isci.2021.102222_bib43) 2018; 12 Wan (10.1016/j.isci.2021.102222_bib57) 2020 Frenkel (10.1016/j.isci.2021.102222_bib10) 2019; 13 Feldmann (10.1016/j.isci.2021.102222_bib9) 2019; 569 Bellec (10.1016/j.isci.2021.102222_bib1) 2020; 11 Serb (10.1016/j.isci.2021.102222_bib53) 2016; 7 Neftci (10.1016/j.isci.2021.102222_bib36) 2019; 36 Masquelier (10.1016/j.isci.2021.102222_bib28) 2007; 3 Zamarreño-Ramos (10.1016/j.isci.2021.102222_bib26) 2011; 5 Scellier (10.1016/j.isci.2021.102222_bib49) 2018 Falez (10.1016/j.isci.2021.102222_bib8) 2019; 93 Davies (10.1016/j.isci.2021.102222_bib5) 2018; 38 Huh (10.1016/j.isci.2021.102222_bib15) 2018 Zhang (10.1016/j.isci.2021.102222_bib60) 2020; 3 Neftci (10.1016/j.isci.2021.102222_bib35) 2017; 11 Hirtzlin (10.1016/j.isci.2021.102222_bib13) 2019 Kheradpisheh (10.1016/j.isci.2021.102222_bib21) 2020; 30 O’Connor (10.1016/j.isci.2021.102222_bib38) 2016 Marković (10.1016/j.isci.2021.102222_bib27) 2020 Bengio (10.1016/j.isci.2021.102222_bib2) 2017; 29 Bichler (10.1016/j.isci.2021.102222_bib4) 2012; 32 Laborieux (10.1016/j.isci.2021.102222_bib23) 2021; 15 Schemmel (10.1016/j.isci.2021.102222_bib51) 2010 Navarro (10.1016/j.isci.2021.102222_bib34) 2020; 9 Thiele (10.1016/j.isci.2021.102222_bib56) 2019 Wang (10.1016/j.isci.2021.102222_bib58) 2020; 5 Ernoult (10.1016/j.isci.2021.102222_bib6) 2019 (10.1016/j.isci.2021.102222_bib29) 1989 Merolla (10.1016/j.isci.2021.102222_bib30) 2014; 345 Ishii (10.1016/j.isci.2021.102222_bib16) 2019 Milo (10.1016/j.isci.2021.102222_bib32) 2020; 13 Kendall (10.1016/j.isci.2021.102222_bib20) 2020 Jo (10.1016/j.isci.2021.102222_bib17) 2010; 10 Zoppo (10.1016/j.isci.2021.102222_bib61) 2020; 14 Rastegari (10.1016/j.isci.2021.102222_bib46) 2016 Kaiser (10.1016/j.isci.2021.102222_bib19) 2020; 14 Mesnard (10.1016/j.isci.2021.102222_bib31) 2016 Park (10.1016/j.isci.2021.102222_bib39) 2020; 55 Scellier (10.1016/j.isci.2021.102222_bib50) 2017; 11 Lee (10.1016/j.isci.2021.102222_bib24) 2016; 10 O’Connor (10.1016/j.isci.2021.102222_bib37) 2019 Furber (10.1016/j.isci.2021.102222_bib11) 2014; 102 Sebastian (10.1016/j.isci.2021.102222_bib52) 2020; 15 Hubara (10.1016/j.isci.2021.102222_bib14) 2016 Sacramento (10.1016/j.isci.2021.102222_bib48) 2018; 31 10.1016/j.isci.2021.102222_bib18 Thakur (10.1016/j.isci.2021.102222_bib55) 2018; 12 Gerstner (10.1016/j.isci.2021.102222_bib12) 2014 Qiao (10.1016/j.isci.2021.102222_bib45) 2015; 9 Prezioso (10.1016/j.isci.2021.102222_bib44) 2018; 9 Bi (10.1016/j.isci.2021.102222_bib3) 2001; 24 Pedretti (10.1016/j.isci.2021.102222_bib42) 2017; 7 Xi (10.1016/j.isci.2021.102222_bib59) 2020; 109 Payvand (10.1016/j.isci.2021.102222_bib41) 2020 Kirkwood (10.1016/j.isci.2021.102222_bib22) 1996; 381 Mostafa (10.1016/j.isci.2021.102222_bib33) 2018; 29 Richards (10.1016/j.isci.2021.102222_bib47) 2019; 22 Li (10.1016/j.isci.2021.102222_bib25) 2015; 106 Tavanaei (10.1016/j.isci.2021.102222_bib54) 2019; 330 Payeur (10.1016/j.isci.2021.102222_bib40) 2020
References_xml	– volume: 10 start-page: 508 year: 2016 ident: bib24 article-title: Training deep spiking neural networks using backpropagation publication-title: Front. Neurosci. – volume: 15 start-page: 1 year: 2020 end-page: 16 ident: bib52 article-title: Memory devices and applications for in-memory computing publication-title: Nat. Nanotechnology – volume: 330 start-page: 39 year: 2019 end-page: 47 ident: bib54 article-title: BP-STDP: approximating backpropagation using spike timing dependent plasticity publication-title: Neurocomputing – start-page: 14.2.1 year: 2019 end-page: 14.2.4 ident: bib16 article-title: On-chip trainable 1.4M 6T2R PCM synaptic array with 1.6K stochastic LIF neurons for spiking RBM publication-title: 2019 IEEE International Electron Devices Meeting (IEDM) – volume: 14 start-page: 424 year: 2020 ident: bib19 article-title: Synaptic plasticity dynamics for deep continuous local learning (DECOLLE) publication-title: Front. Neurosci. – volume: 5 start-page: 173 year: 2020 end-page: 195 ident: bib58 article-title: Resistive switching materials for information processing publication-title: Nat. Rev. Mater. – volume: 569 start-page: 208 year: 2019 end-page: 214 ident: bib9 article-title: All-optical spiking neurosynaptic networks with self-learning capabilities publication-title: Nature – start-page: 7081 year: 2019 end-page: 7091 ident: bib6 article-title: Updates of equilibrium prop match gradients of backprop through time in an RNN with static input publication-title: Advances in Neural Information Processing Systems 32 – year: 2016 ident: bib38 article-title: Deep spiking networks publication-title: arXiv – volume: 15 start-page: 129 year: 2021 ident: bib23 article-title: Scaling equilibrium propagation to deep ConvNets by drastically reducing its gradient estimator bias publication-title: Frontiers in Neuroscience – volume: 13 start-page: 145 year: 2019 end-page: 158 ident: bib10 article-title: A 0.086-mm publication-title: IEEE Trans. Biomed. Circuits Syst. – volume: 55 start-page: 108 year: 2020 end-page: 119 ident: bib39 article-title: A 65-nm neuromorphic image classification processor with energy-efficient training through direct spike-only feedback publication-title: IEEE J. Solid-State Circuits – start-page: 4107 year: 2016 end-page: 4115 ident: bib14 article-title: Binarized neural networks publication-title: Advances in Neural Information Processing Systems 29 – volume: 30 start-page: 2050027 year: 2020 ident: bib21 article-title: Temporal backpropagation for spiking neural networks with one spike per neuron publication-title: Int. J. Neural Syst. – year: 1989 ident: bib29 publication-title: Analog VLSI Implementation of Neural Systems – year: 2014 ident: bib12 article-title: Neuronal Dynamics: From Single Neurons to Networks and Models of Cognition – year: 2020 ident: bib7 article-title: Equilibrium Propagation with Continual Weight Updates publication-title: arXiv – volume: 31 start-page: 8721 year: 2018 end-page: 8732 ident: bib48 article-title: Dendritic cortical microcircuits approximate the backpropagation algorithm publication-title: Advances in Neural Information Processing Systems – volume: 29 start-page: 555 year: 2017 end-page: 577 ident: bib2 article-title: STDP-compatible approximation of backpropagation in an energy-based model publication-title: Neural Comput. – start-page: 22.6.1 year: 2019 end-page: 22.6.4 ident: bib13 article-title: Hybrid analog-digital learning with differential RRAM synapses publication-title: 2019 IEEE International Electron Devices Meeting (IEDM). 2019 IEEE International Electron Devices Meeting (IEDM) – volume: 12 start-page: 991 year: 2018 ident: bib55 article-title: Large-scale neuromorphic spiking array processors: a quest to mimic the brain publication-title: Front. Neurosci. – volume: 106 start-page: 212902 year: 2015 ident: bib25 article-title: High-endurance megahertz electrical self-oscillation in Ti/NbOx bilayer structures publication-title: Appl. Phys. Lett. – volume: 9 start-page: e54940 year: 2020 ident: bib34 article-title: Sodium channels implement a molecular leaky integrator that detects action potentials and regulates neuronal firing publication-title: eLife – year: 2018 ident: bib49 article-title: Generalization of Equilibrium Propagation to Vector Field Dynamics publication-title: arXiv – volume: 3 start-page: 371 year: 2020 end-page: 382 ident: bib60 article-title: Neuro-inspired computing chips publication-title: Nat. Electronics – start-page: 1947 year: 2010 end-page: 1950 ident: bib51 article-title: A wafer-scale neuromorphic hardware system for large-scale neural modeling publication-title: Proceedings of 2010 IEEE International Symposium on Circuits and Systems – volume: 36 start-page: 51 year: 2019 end-page: 63 ident: bib36 article-title: Surrogate gradient learning in spiking neural networks: bringing the power of gradient-based optimization to spiking neural networks publication-title: IEEE Signal Process. Mag. – volume: 13 start-page: 166 year: 2020 ident: bib32 article-title: Memristive and CMOS devices for neuromorphic computing publication-title: Materials – year: 2016 ident: bib31 article-title: Towards deep learning with spiking neurons in energy based models with contrastive Hebbian plasticity publication-title: arXiv – volume: 9 start-page: 141 year: 2015 ident: bib45 article-title: A reconfigurable on-line learning spiking neuromorphic processor comprising 256 neurons and 128K synapses publication-title: Front. Neurosci. – volume: 7 start-page: 5288 year: 2017 ident: bib42 article-title: Memristive neural network for on-line learning and tracking with brain-inspired spike timing dependent plasticity publication-title: Sci. Rep. – year: 2019 ident: bib56 article-title: SpikeGrad: an ANN-equivalent computation model for implementing backpropagation with spikes publication-title: arXiv – volume: 24 start-page: 139 year: 2001 end-page: 166 ident: bib3 article-title: Synaptic modification by correlated activity: hebb’s postulate revisited publication-title: Annu. Rev. Neurosci. – volume: 32 start-page: 339 year: 2012 end-page: 348 ident: bib4 article-title: Extraction of temporally correlated features from dynamic vision sensors with spike-timing-dependent plasticity publication-title: Neural Networks – start-page: 1 year: 2020 end-page: 12 ident: bib27 article-title: Physics for neuromorphic computing publication-title: Nat. Rev. Phys. – volume: 11 start-page: 24 year: 2017 ident: bib50 article-title: Equilibrium propagation: bridging the gap between energy-based models and backpropagation publication-title: Front. Comput. Neurosci. – volume: 10 start-page: 1297 year: 2010 end-page: 1301 ident: bib17 article-title: Nanoscale memristor device as synapse in neuromorphic systems publication-title: Nano Lett. – volume: 381 start-page: 526 year: 1996 end-page: 528 ident: bib22 article-title: Experience-dependent modification of synaptic plasticity in visual cortex publication-title: Nature – volume: 7 year: 2016 ident: bib53 article-title: Unsupervised learning in probabilistic neural networks with multi-state metal-oxide memristive synapses publication-title: Nat. Commun. – volume: 93 start-page: 418 year: 2019 end-page: 429 ident: bib8 article-title: Unsupervised visual feature learning with spike-timing-dependent plasticity: how far are we from traditional feature learning approaches? publication-title: Pattern Recognit. – year: 2020 ident: bib40 article-title: Burst-dependent synaptic plasticity can coordinate learning in hierarchical circuits publication-title: bioRxiv – start-page: 1516 year: 2019 end-page: 1523 ident: bib37 article-title: Training a spiking neural network with equilibrium propagation publication-title: The 22nd International Conference on Artificial Intelligence and Statistics. The 22nd International Conference on Artificial Intelligence and Statistics – volume: 3 start-page: e31 year: 2007 ident: bib28 article-title: Unsupervised learning of visual features through spike timing dependent plasticity publication-title: PLoS Comput. Biol. – volume: 9 start-page: 5311 year: 2018 ident: bib44 article-title: Spike-timing-dependent plasticity learning of coincidence detection with passively integrated memristive circuits publication-title: Nat. Commun. – volume: 11 start-page: 1621 year: 2017 end-page: 1671 ident: bib35 article-title: Event-driven random back-propagation: enabling neuromorphic deep learning machines publication-title: Front. Neurosci. – volume: 11 start-page: 3625 year: 2020 ident: bib1 article-title: A solution to the learning dilemma for recurrent networks of spiking neurons publication-title: Nat. Commun. – year: 2020 ident: bib20 article-title: Training end-to-end analog neural networks with equilibrium propagation publication-title: arXiv – volume: 29 start-page: 3227 year: 2018 end-page: 3235 ident: bib33 article-title: Supervised learning based on temporal coding in spiking neural networks publication-title: IEEE Trans. Neural Networks Learn. Syst. – volume: 102 start-page: 652 year: 2014 end-page: 665 ident: bib11 article-title: The SpiNNaker project publication-title: Proc. IEEE – volume: 22 start-page: 1761 year: 2019 end-page: 1770 ident: bib47 article-title: A deep learning framework for neuroscience publication-title: Nat. Neurosci. – volume: 38 start-page: 82 year: 2018 end-page: 99 ident: bib5 article-title: Loihi: a neuromorphic manycore processor with on-chip learning publication-title: IEEE Micro – start-page: 1433 year: 2018 end-page: 1443 ident: bib15 article-title: Gradient descent for spiking neural networks publication-title: Advances in Neural Information Processing Systems 31 – start-page: 525 year: 2016 end-page: 542 ident: bib46 article-title: XNOR-net: ImageNet classification using binary convolutional neural networks publication-title: Computer Vision – ECCV 2016 – start-page: 498 year: 2020 end-page: 500 ident: bib57 article-title: 33.1 A 74 TMACS/W CMOS-RRAM neurosynaptic core with dynamically reconfigurable dataflow and in-situ transposable weights for probabilistic graphical models publication-title: 2020 IEEE International Solid- State Circuits Conference - (ISSCC). 2020 – volume: 14 start-page: 240 year: 2020 ident: bib61 article-title: Equilibrium propagation for memristor-based recurrent neural networks publication-title: Front. Neurosci. – volume: 5 start-page: 26 year: 2011 ident: bib26 article-title: On spike-timing-dependent-plasticity, memristive devices, and building a self-learning visual cortex publication-title: Front. Neurosci. – start-page: 218 year: 2020 end-page: 222 ident: bib41 article-title: Error-triggered three-factor learning dynamics for crossbar arrays publication-title: 2020 2nd IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) – volume: 12 start-page: 126 year: 2018 ident: bib43 article-title: Deep learning with spiking neurons: opportunities and challenges publication-title: Front. Neurosci. – reference: Joseph, V. and Nagarajan, C. (2020 )MADONNA: A Framework for Energy Measurements and Assistance in Designing Low Power Deep Neural Networks. p. 7. – volume: 109 start-page: 1 year: 2020 end-page: 29 ident: bib59 article-title: In-memory learning with analog resistive switching memory: a review and perspective publication-title: Proc. IEEE – volume: 345 start-page: 668 year: 2014 end-page: 673 ident: bib30 article-title: A million spiking-neuron integrated circuit with a scalable communication network and interface publication-title: Science – year: 2016 ident: 10.1016/j.isci.2021.102222_bib31 article-title: Towards deep learning with spiking neurons in energy based models with contrastive Hebbian plasticity publication-title: arXiv – volume: 9 start-page: 141 year: 2015 ident: 10.1016/j.isci.2021.102222_bib45 article-title: A reconfigurable on-line learning spiking neuromorphic processor comprising 256 neurons and 128K synapses publication-title: Front. Neurosci. doi: 10.3389/fnins.2015.00141 – start-page: 14.2.1 year: 2019 ident: 10.1016/j.isci.2021.102222_bib16 article-title: On-chip trainable 1.4M 6T2R PCM synaptic array with 1.6K stochastic LIF neurons for spiking RBM – volume: 15 start-page: 129 year: 2021 ident: 10.1016/j.isci.2021.102222_bib23 article-title: Scaling equilibrium propagation to deep ConvNets by drastically reducing its gradient estimator bias publication-title: Frontiers in Neuroscience doi: 10.3389/fnins.2021.633674 – ident: 10.1016/j.isci.2021.102222_bib18 – volume: 14 start-page: 424 year: 2020 ident: 10.1016/j.isci.2021.102222_bib19 article-title: Synaptic plasticity dynamics for deep continuous local learning (DECOLLE) publication-title: Front. Neurosci. doi: 10.3389/fnins.2020.00424 – volume: 11 start-page: 1621 year: 2017 ident: 10.1016/j.isci.2021.102222_bib35 article-title: Event-driven random back-propagation: enabling neuromorphic deep learning machines publication-title: Front. Neurosci. doi: 10.3389/fnins.2017.00324 – volume: 93 start-page: 418 year: 2019 ident: 10.1016/j.isci.2021.102222_bib8 article-title: Unsupervised visual feature learning with spike-timing-dependent plasticity: how far are we from traditional feature learning approaches? publication-title: Pattern Recognit. doi: 10.1016/j.patcog.2019.04.016 – volume: 5 start-page: 173 year: 2020 ident: 10.1016/j.isci.2021.102222_bib58 article-title: Resistive switching materials for information processing publication-title: Nat. Rev. Mater. doi: 10.1038/s41578-019-0159-3 – year: 2018 ident: 10.1016/j.isci.2021.102222_bib49 article-title: Generalization of Equilibrium Propagation to Vector Field Dynamics publication-title: arXiv – volume: 3 start-page: 371 year: 2020 ident: 10.1016/j.isci.2021.102222_bib60 article-title: Neuro-inspired computing chips publication-title: Nat. Electronics doi: 10.1038/s41928-020-0435-7 – start-page: 525 year: 2016 ident: 10.1016/j.isci.2021.102222_bib46 article-title: XNOR-net: ImageNet classification using binary convolutional neural networks – volume: 7 year: 2016 ident: 10.1016/j.isci.2021.102222_bib53 article-title: Unsupervised learning in probabilistic neural networks with multi-state metal-oxide memristive synapses publication-title: Nat. Commun. doi: 10.1038/ncomms12611 – volume: 109 start-page: 1 year: 2020 ident: 10.1016/j.isci.2021.102222_bib59 article-title: In-memory learning with analog resistive switching memory: a review and perspective publication-title: Proc. IEEE – volume: 29 start-page: 555 year: 2017 ident: 10.1016/j.isci.2021.102222_bib2 article-title: STDP-compatible approximation of backpropagation in an energy-based model publication-title: Neural Comput. doi: 10.1162/NECO_a_00934 – start-page: 1947 year: 2010 ident: 10.1016/j.isci.2021.102222_bib51 article-title: A wafer-scale neuromorphic hardware system for large-scale neural modeling – volume: 12 start-page: 126 year: 2018 ident: 10.1016/j.isci.2021.102222_bib43 article-title: Deep learning with spiking neurons: opportunities and challenges publication-title: Front. Neurosci. doi: 10.3389/fnins.2018.00774 – volume: 10 start-page: 1297 year: 2010 ident: 10.1016/j.isci.2021.102222_bib17 article-title: Nanoscale memristor device as synapse in neuromorphic systems publication-title: Nano Lett. doi: 10.1021/nl904092h – volume: 32 start-page: 339 year: 2012 ident: 10.1016/j.isci.2021.102222_bib4 article-title: Extraction of temporally correlated features from dynamic vision sensors with spike-timing-dependent plasticity publication-title: Neural Networks doi: 10.1016/j.neunet.2012.02.022 – volume: 569 start-page: 208 year: 2019 ident: 10.1016/j.isci.2021.102222_bib9 article-title: All-optical spiking neurosynaptic networks with self-learning capabilities publication-title: Nature doi: 10.1038/s41586-019-1157-8 – start-page: 1516 year: 2019 ident: 10.1016/j.isci.2021.102222_bib37 article-title: Training a spiking neural network with equilibrium propagation – volume: 30 start-page: 2050027 year: 2020 ident: 10.1016/j.isci.2021.102222_bib21 article-title: Temporal backpropagation for spiking neural networks with one spike per neuron publication-title: Int. J. Neural Syst. doi: 10.1142/S0129065720500276 – volume: 5 start-page: 26 year: 2011 ident: 10.1016/j.isci.2021.102222_bib26 article-title: On spike-timing-dependent-plasticity, memristive devices, and building a self-learning visual cortex publication-title: Front. Neurosci. doi: 10.3389/fnins.2011.00026 – volume: 9 start-page: 5311 year: 2018 ident: 10.1016/j.isci.2021.102222_bib44 article-title: Spike-timing-dependent plasticity learning of coincidence detection with passively integrated memristive circuits publication-title: Nat. Commun. doi: 10.1038/s41467-018-07757-y – volume: 9 start-page: e54940 year: 2020 ident: 10.1016/j.isci.2021.102222_bib34 article-title: Sodium channels implement a molecular leaky integrator that detects action potentials and regulates neuronal firing publication-title: eLife doi: 10.7554/eLife.54940 – volume: 31 start-page: 8721 year: 2018 ident: 10.1016/j.isci.2021.102222_bib48 article-title: Dendritic cortical microcircuits approximate the backpropagation algorithm – start-page: 218 year: 2020 ident: 10.1016/j.isci.2021.102222_bib41 article-title: Error-triggered three-factor learning dynamics for crossbar arrays – start-page: 498 year: 2020 ident: 10.1016/j.isci.2021.102222_bib57 article-title: 33.1 A 74 TMACS/W CMOS-RRAM neurosynaptic core with dynamically reconfigurable dataflow and in-situ transposable weights for probabilistic graphical models – start-page: 22.6.1 year: 2019 ident: 10.1016/j.isci.2021.102222_bib13 article-title: Hybrid analog-digital learning with differential RRAM synapses – start-page: 4107 year: 2016 ident: 10.1016/j.isci.2021.102222_bib14 article-title: Binarized neural networks – volume: 14 start-page: 240 year: 2020 ident: 10.1016/j.isci.2021.102222_bib61 article-title: Equilibrium propagation for memristor-based recurrent neural networks publication-title: Front. Neurosci. doi: 10.3389/fnins.2020.00240 – volume: 7 start-page: 5288 year: 2017 ident: 10.1016/j.isci.2021.102222_bib42 article-title: Memristive neural network for on-line learning and tracking with brain-inspired spike timing dependent plasticity publication-title: Sci. Rep. doi: 10.1038/s41598-017-05480-0 – year: 2019 ident: 10.1016/j.isci.2021.102222_bib56 article-title: SpikeGrad: an ANN-equivalent computation model for implementing backpropagation with spikes publication-title: arXiv – volume: 102 start-page: 652 year: 2014 ident: 10.1016/j.isci.2021.102222_bib11 article-title: The SpiNNaker project publication-title: Proc. IEEE doi: 10.1109/JPROC.2014.2304638 – start-page: 1 year: 2020 ident: 10.1016/j.isci.2021.102222_bib27 article-title: Physics for neuromorphic computing publication-title: Nat. Rev. Phys. – volume: 24 start-page: 139 year: 2001 ident: 10.1016/j.isci.2021.102222_bib3 article-title: Synaptic modification by correlated activity: hebb’s postulate revisited publication-title: Annu. Rev. Neurosci. doi: 10.1146/annurev.neuro.24.1.139 – year: 2016 ident: 10.1016/j.isci.2021.102222_bib38 article-title: Deep spiking networks publication-title: arXiv – volume: 345 start-page: 668 year: 2014 ident: 10.1016/j.isci.2021.102222_bib30 article-title: A million spiking-neuron integrated circuit with a scalable communication network and interface publication-title: Science doi: 10.1126/science.1254642 – volume: 29 start-page: 3227 issue: 7 year: 2018 ident: 10.1016/j.isci.2021.102222_bib33 article-title: Supervised learning based on temporal coding in spiking neural networks publication-title: IEEE Trans. Neural Networks Learn. Syst. – year: 2020 ident: 10.1016/j.isci.2021.102222_bib7 article-title: Equilibrium Propagation with Continual Weight Updates publication-title: arXiv – volume: 12 start-page: 991 year: 2018 ident: 10.1016/j.isci.2021.102222_bib55 article-title: Large-scale neuromorphic spiking array processors: a quest to mimic the brain publication-title: Front. Neurosci. doi: 10.3389/fnins.2018.00991 – volume: 13 start-page: 166 year: 2020 ident: 10.1016/j.isci.2021.102222_bib32 article-title: Memristive and CMOS devices for neuromorphic computing publication-title: Materials doi: 10.3390/ma13010166 – volume: 11 start-page: 24 year: 2017 ident: 10.1016/j.isci.2021.102222_bib50 article-title: Equilibrium propagation: bridging the gap between energy-based models and backpropagation publication-title: Front. Comput. Neurosci. doi: 10.3389/fncom.2017.00024 – start-page: 7081 year: 2019 ident: 10.1016/j.isci.2021.102222_bib6 article-title: Updates of equilibrium prop match gradients of backprop through time in an RNN with static input – year: 2020 ident: 10.1016/j.isci.2021.102222_bib40 article-title: Burst-dependent synaptic plasticity can coordinate learning in hierarchical circuits publication-title: bioRxiv – volume: 13 start-page: 145 issue: 1 year: 2019 ident: 10.1016/j.isci.2021.102222_bib10 article-title: A 0.086-mm2 12.7-pJ/SOP 64k-synapse 256-neuron online-learning digital spiking neuromorphic processor in 28-nm CMOS publication-title: IEEE Trans. Biomed. Circuits Syst. – start-page: 1433 year: 2018 ident: 10.1016/j.isci.2021.102222_bib15 article-title: Gradient descent for spiking neural networks – volume: 381 start-page: 526 year: 1996 ident: 10.1016/j.isci.2021.102222_bib22 article-title: Experience-dependent modification of synaptic plasticity in visual cortex publication-title: Nature doi: 10.1038/381526a0 – volume: 330 start-page: 39 year: 2019 ident: 10.1016/j.isci.2021.102222_bib54 article-title: BP-STDP: approximating backpropagation using spike timing dependent plasticity publication-title: Neurocomputing doi: 10.1016/j.neucom.2018.11.014 – volume: 36 start-page: 51 year: 2019 ident: 10.1016/j.isci.2021.102222_bib36 article-title: Surrogate gradient learning in spiking neural networks: bringing the power of gradient-based optimization to spiking neural networks publication-title: IEEE Signal Process. Mag. doi: 10.1109/MSP.2019.2931595 – volume: 22 start-page: 1761 year: 2019 ident: 10.1016/j.isci.2021.102222_bib47 article-title: A deep learning framework for neuroscience publication-title: Nat. Neurosci. doi: 10.1038/s41593-019-0520-2 – year: 2020 ident: 10.1016/j.isci.2021.102222_bib20 article-title: Training end-to-end analog neural networks with equilibrium propagation publication-title: arXiv – volume: 55 start-page: 108 year: 2020 ident: 10.1016/j.isci.2021.102222_bib39 article-title: A 65-nm neuromorphic image classification processor with energy-efficient training through direct spike-only feedback publication-title: IEEE J. Solid-State Circuits doi: 10.1109/JSSC.2019.2942367 – volume: 15 start-page: 1 year: 2020 ident: 10.1016/j.isci.2021.102222_bib52 article-title: Memory devices and applications for in-memory computing publication-title: Nat. Nanotechnology – year: 2014 ident: 10.1016/j.isci.2021.102222_bib12 – year: 1989 ident: 10.1016/j.isci.2021.102222_bib29 – volume: 10 start-page: 508 year: 2016 ident: 10.1016/j.isci.2021.102222_bib24 article-title: Training deep spiking neural networks using backpropagation publication-title: Front. Neurosci. doi: 10.3389/fnins.2016.00508 – volume: 38 start-page: 82 year: 2018 ident: 10.1016/j.isci.2021.102222_bib5 article-title: Loihi: a neuromorphic manycore processor with on-chip learning publication-title: IEEE Micro doi: 10.1109/MM.2018.112130359 – volume: 106 start-page: 212902 year: 2015 ident: 10.1016/j.isci.2021.102222_bib25 article-title: High-endurance megahertz electrical self-oscillation in Ti/NbOx bilayer structures publication-title: Appl. Phys. Lett. doi: 10.1063/1.4921745 – volume: 3 start-page: e31 year: 2007 ident: 10.1016/j.isci.2021.102222_bib28 article-title: Unsupervised learning of visual features through spike timing dependent plasticity publication-title: PLoS Comput. Biol. doi: 10.1371/journal.pcbi.0030031 – volume: 11 start-page: 3625 year: 2020 ident: 10.1016/j.isci.2021.102222_bib1 article-title: A solution to the learning dilemma for recurrent networks of spiking neurons publication-title: Nat. Commun. doi: 10.1038/s41467-020-17236-y
SSID	ssj0002002496
Score	2.4311783
Snippet	Finding spike-based learning algorithms that can be implemented within the local constraints of neuromorphic systems, while achieving high accuracy, remains a...
SourceID	doaj pubmedcentral hal proquest pubmed crossref elsevier
SourceType	Open Website Open Access Repository Aggregation Database Index Database Enrichment Source Publisher
StartPage	102222
SubjectTerms	Algorithms Artificial Intelligence Computer Science Physics
Title	EqSpike: Spike-driven equilibrium propagation for neuromorphic implementations
URI	https://dx.doi.org/10.1016/j.isci.2021.102222 https://www.ncbi.nlm.nih.gov/pubmed/33748709 https://www.proquest.com/docview/2503682050 https://hal.science/hal-03451693 https://pubmed.ncbi.nlm.nih.gov/PMC7970361 https://doaj.org/article/15fecf18376346d68c50bc97ba8803cd
Volume	24
WOSCitedRecordID	wos000631646000072&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVAON databaseName: DOAJ Directory of Open Access Journals customDbUrl: eissn: 2589-0042 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002002496 issn: 2589-0042 databaseCode: DOA dateStart: 20180101 isFulltext: true titleUrlDefault: https://www.doaj.org/ providerName: Directory of Open Access Journals
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3LbtQwFLVoxYIN70cKVAGxQwE7tuOYXUGtukAjJECaneVX1BQm005n-v2910lGE5DKhlUk52HF99g-Vzk5l5B3WMxdhVoX1leuEM7HwgKzKGjjBNMK0jHZF5tQs1k9n-tvO6W-UBPW2wP3A_eRySb6BoAHE0FUoaq9pM5r5Swgj_uAqy9VeieZOk-f19AKL1WWk6gJAmgOf8z04i784xWSw5J9SBlPOdmVknn_ZHPaO0OV5N8U9E8l5c7WdPKQ3B84ZX7Uv8sjcid2j8mDsV5DPkzfJ2R2fPn9ov0VP-XpUIQVLnV5vNy0Sfm_WeTQN6wwKVo50Nk82V0ulhCM1uftYhSbJ7A-JT9Pjn98OS2GegqFl1KuCx6F46osvW0ChzwFuEfdaOq9qrwsNa2DDAEIXPSi5o2wynoWmBccx5zRwJ-R_W7ZxRckr2xTKeUaSNa4cI7ZykoNTUK7mnvaZISN42n8YDaONS9-m1FVdm4wBgZjYPoYZOT99p6L3mrj1qs_Y5i2V6JNdmoA8JgBPOZf4MmIHINsBsbRMwl4VHtr528BEZO-T4--GmyjHEsfa37NMvJmBIyBOYsfYmwXl5srA7STV0C9JM3I8x5A22dx9ANSVGdETaA16Wx6pmvPki-40uimxg7-x8i8JPfwfVFtx_Qrsr9ebeJrctdfr9ur1SHZU_P6ME25G_rnLY8
linkProvider	Directory of Open Access Journals
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=EqSpike%3A+spike-driven+equilibrium+propagation+for+neuromorphic+implementations&rft.jtitle=iScience&rft.au=Martin%2C+Erwann&rft.au=Ernoult%2C+Maxence&rft.au=Laydevant%2C+J%C3%A9r%C3%A9mie&rft.au=Li%2C+Shuai&rft.date=2021-03-19&rft.eissn=2589-0042&rft.volume=24&rft.issue=3&rft.spage=102222&rft_id=info:doi/10.1016%2Fj.isci.2021.102222&rft_id=info%3Apmid%2F33748709&rft.externalDocID=33748709
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2589-0042&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2589-0042&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2589-0042&client=summon