{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T17:44:29Z","timestamp":1780595069085,"version":"3.54.1"},"reference-count":43,"publisher":"IOP Publishing","issue":"3","license":[{"start":{"date-parts":[[2025,7,25]],"date-time":"2025-07-25T00:00:00Z","timestamp":1753401600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,7,25]],"date-time":"2025-07-25T00:00:00Z","timestamp":1753401600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"name":"European Research Council","award":["101020684"],"award-info":[{"award-number":["101020684"]}]},{"DOI":"10.13039\/501100001665","name":"Agence Nationale de la Recherche","doi-asserted-by":"crossref","award":["ANR-23-PEIA-0002"],"award-info":[{"award-number":["ANR-23-PEIA-0002"]}],"id":[{"id":"10.13039\/501100001665","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Neuromorph. Comput. Eng."],"published-print":{"date-parts":[[2025,9,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Oscillator networks represent a promising technology for unconventional computing and artificial intelligence. Thus far, these systems have primarily been demonstrated in small-scale implementations, such as Ising machines for solving combinatorial problems and associative memories for image recognition, typically trained without state-of-the-art gradient-based algorithms. Scaling up oscillator-based systems requires advanced gradient-based training methods that also ensure robustness against frequency dispersion between individual oscillators. Here, we demonstrate through simulations that the Equilibrium Propagation algorithm enables effective gradient-based training of oscillator networks, facilitating synchronization even when initial oscillator frequencies are significantly dispersed. We specifically investigate two oscillator models: purely phase-coupled oscillators and oscillators coupled via both amplitude and phase interactions. Our results show that these oscillator networks can scale successfully to standard image recognition benchmarks, such as achieving nearly 98% test accuracy on the Modified National Institute of Standards and Technology (MNIST) dataset, despite noise introduced by imperfect synchronization. This work thus paves the way for practical hardware implementations of large-scale oscillator networks, such as those based on spintronic devices.<\/jats:p>","DOI":"10.1088\/2634-4386\/adebaa","type":"journal-article","created":{"date-parts":[[2025,7,3]],"date-time":"2025-07-03T18:50:21Z","timestamp":1751568621000},"page":"034008","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Training and synchronizing oscillator networks with Equilibrium Propagation"],"prefix":"10.1088","volume":"5","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-0881-2917","authenticated-orcid":true,"given":"Th\u00e9ophile","family":"Rageau","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4866-4490","authenticated-orcid":false,"given":"Julie","family":"Grollier","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2025,7,25]]},"reference":[{"key":"nceadebaabib1","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s44335-024-00015-z","article-title":"Computing with oscillators from theoretical underpinnings to applications and demonstrators","volume":"1","author":"Todri-Sanial","year":"2024","journal-title":"npj Unconv. Comput."},{"key":"nceadebaabib2","first-page":"pp 232","article-title":"OIM: oscillator-based Ising machines for solving combinatorial optimisation problems","author":"Wang","year":"2019"},{"key":"nceadebaabib3","doi-asserted-by":"publisher","first-page":"310","DOI":"10.1038\/s41928-022-00749-3","article-title":"A 1,968-node coupled ring oscillator circuit for combinatorial optimization problem solving","volume":"5","author":"Moy","year":"2022","journal-title":"Nat. Electron."},{"key":"nceadebaabib4","first-page":"pp 602","article-title":"Optimizing with phases: design and application space assessment for networks of phase-locked ring oscillators","author":"Bazzi","year":"2024"},{"key":"nceadebaabib5","doi-asserted-by":"publisher","first-page":"333","DOI":"10.1038\/s42005-022-01111-x","article-title":"An Ising machine based on networks of subharmonic electrical resonators","volume":"5","author":"English","year":"2022","journal-title":"Commun. Phys."},{"key":"nceadebaabib6","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-019-49699-5","article-title":"Analog coupled oscillator based weighted Ising machine","volume":"9","author":"Chou","year":"2019","journal-title":"Sci. Rep."},{"key":"nceadebaabib7","doi-asserted-by":"publisher","DOI":"10.1088\/2634-4386\/ace9f5","article-title":"A mixed-signal oscillatory neural network for scalable analog computations in phase domain","volume":"3","author":"Delacour","year":"2023","journal-title":"Neuromorph. Comput. Eng."},{"key":"nceadebaabib8","doi-asserted-by":"publisher","first-page":"116","DOI":"10.1038\/s44172-024-00261-w","article-title":"An integrated coupled oscillator network to solve optimization problems","volume":"3","author":"Graber","year":"2024","journal-title":"Commun. Eng."},{"key":"nceadebaabib9","article-title":"A neural network of weakly coupled nonlinear oscillators with a global, time-dependent coupling-theory and experiment","author":"H\u00f6lzel","year":"2013"},{"key":"nceadebaabib10","doi-asserted-by":"publisher","first-page":"170","DOI":"10.1109\/JXCDC.2020.3046143","article-title":"Convolution inference via synchronization of a coupled CMOS oscillator array","volume":"6","author":"Nikonov","year":"2020","journal-title":"IEEE J. Exploratory Solid State Comput. Dev. Circuits"},{"key":"nceadebaabib11","doi-asserted-by":"crossref","DOI":"10.1109\/IMW61990.2025.11026987","article-title":"Hardware implementation of ring oscillator networks coupled by BEOL integrated ReRAM for associative memory tasks","author":"Choi","year":"2025"},{"key":"nceadebaabib12","doi-asserted-by":"publisher","first-page":"428","DOI":"10.1038\/nature23011","article-title":"Neuromorphic computing with nanoscale spintronic oscillators","volume":"547","author":"Torrejon","year":"2017","journal-title":"Nature"},{"key":"nceadebaabib13","doi-asserted-by":"publisher","first-page":"230","DOI":"10.1038\/s41586-018-0632-y","article-title":"Vowel recognition with four coupled spin-torque nano-oscillators","volume":"563","author":"Romera","year":"2018","journal-title":"Nature"},{"key":"nceadebaabib14","author":"Buzsaki","year":"2006"},{"key":"nceadebaabib15","doi-asserted-by":"publisher","first-page":"102","DOI":"10.1038\/scientificamerican1293-102","article-title":"Coupled oscillators and biological synchronization","volume":"269","author":"Strogatz","year":"1993","journal-title":"Sci. Am."},{"key":"nceadebaabib16","doi-asserted-by":"publisher","first-page":"42","DOI":"10.1016\/j.jneumeth.2009.05.015","article-title":"Synchronization phenomena in human epileptic brain networks","volume":"183","author":"Lehnertz","year":"2009","journal-title":"J. Neurosci. Methods"},{"key":"nceadebaabib17","doi-asserted-by":"publisher","first-page":"97","DOI":"10.3389\/fncom.2015.00097","article-title":"Synchronization-based computation through networks of coupled oscillators","volume":"9","author":"Malagarriga","year":"2015","journal-title":"Front. Comput. Neurosci."},{"key":"nceadebaabib18","doi-asserted-by":"publisher","first-page":"2983","DOI":"10.1103\/PhysRevLett.82.2983","article-title":"Oscillatory neurocomputers with dynamic connectivity","volume":"82","author":"Hoppensteadt","year":"1999","journal-title":"Phys. Rev. Lett."},{"key":"nceadebaabib19","first-page":"pp 420","article-title":"Self-entrainment of a population of coupled non-linear oscillators","author":"Kuramoto","year":"1975"},{"key":"nceadebaabib20","doi-asserted-by":"publisher","DOI":"10.1063\/1.5042423","article-title":"Simulation of coupled spin torque oscillators for pattern recognition","volume":"124","author":"Popescu","year":"2018","journal-title":"J. Appl. Phys."},{"key":"nceadebaabib21","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2024.1307525","article-title":"Design of oscillatory neural networks by machine learning","volume":"18","author":"Rudner","year":"2024","journal-title":"Front. Neurosci."},{"key":"nceadebaabib22","doi-asserted-by":"publisher","first-page":"08","DOI":"10.1109\/TNNLS.2014.2345572","article-title":"Phase oscillatory network and visual pattern recognition","volume":"26","author":"Follmann","year":"2014","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"nceadebaabib23","doi-asserted-by":"publisher","first-page":"533","DOI":"10.1038\/323533a0","article-title":"Learning representations by back-propagating errors","volume":"323","author":"Rumelhart","year":"1986","journal-title":"Nature"},{"key":"nceadebaabib24","doi-asserted-by":"publisher","first-page":"1550","DOI":"10.1109\/5.58337","article-title":"Backpropagation through time: what it does and how to do it","volume":"78","author":"Werbos","year":"1990","journal-title":"Proc. IEEE"},{"key":"nceadebaabib25","doi-asserted-by":"publisher","DOI":"10.1088\/1742-6596\/1804\/1\/012169","article-title":"Backpropagation algorithm and its hardware implementations: A review","volume":"1804","author":"Kuninti","year":"2021","journal-title":"J. Phys.: Conf. Ser."},{"key":"nceadebaabib26","article-title":"Energy-based learning algorithms for analog computing: a comparative study","volume":"vol 36","author":"Scellier","year":"2024"},{"key":"nceadebaabib27","doi-asserted-by":"publisher","first-page":"24","DOI":"10.3389\/fncom.2017.00024","article-title":"Equilibrium propagation: bridging the gap between energy-based models and backpropagation","volume":"11","author":"Scellier","year":"2017","journal-title":"Front. Comput. Neurosci."},{"key":"nceadebaabib28","doi-asserted-by":"publisher","first-page":"45","DOI":"10.1038\/s41928-022-00869-w","article-title":"Activity-difference training of deep neural networks using memristor crossbars","volume":"6","author":"Yi","year":"2023","journal-title":"Nat. Electron."},{"key":"nceadebaabib29","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2319718121","article-title":"Machine learning without a processor: emergent learning in a nonlinear analog network","volume":"121","author":"Dillavou","year":"2024","journal-title":"Proc. Natl Acad. Sci."},{"key":"nceadebaabib30","doi-asserted-by":"publisher","first-page":"3671","DOI":"10.1038\/s41467-024-46879-4","article-title":"Training an Ising machine with equilibrium propagation","volume":"15","author":"Laydevant","year":"2024","journal-title":"Nat. Commun."},{"key":"nceadebaabib31","article-title":"Towards training digitally-tied analog blocks via hybrid gradient computation","author":"Nest","year":"2024"},{"key":"nceadebaabib32","doi-asserted-by":"crossref","DOI":"10.1088\/2634-4386\/ad752b","article-title":"Training coupled phase oscillators as a neuromorphic platform using equilibrium propagation","author":"Wang","year":"2024"},{"key":"nceadebaabib33","doi-asserted-by":"publisher","first-page":"7","DOI":"10.1038\/s43246-023-00423-x","article-title":"The impact of local pinning sites in magnetic tunnel junctions with non-homogeneous free layers","volume":"5","author":"Jenkins","year":"2024","journal-title":"Commun. Mater."},{"key":"nceadebaabib34","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-018-31769-9","article-title":"Scaling up electrically synchronized spin torque oscillator networks","volume":"8","author":"Tsunegi","year":"2018","journal-title":"Sci. Rep."},{"key":"nceadebaabib35","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.physrep.2015.10.008","article-title":"The kuramoto model in complex networks","volume":"610","author":"Rodrigues","year":"2016","journal-title":"Phys. Rep."},{"key":"nceadebaabib36","doi-asserted-by":"publisher","first-page":"1875","DOI":"10.1109\/TMAG.2008.2009935","article-title":"Nonlinear auto-oscillator theory of microwave generation by spin-polarized current","volume":"45","author":"Slavin","year":"2009","journal-title":"IEEE Trans. Magn."},{"key":"nceadebaabib37","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2021.633674","article-title":"Scaling equilibrium propagation to deep convnets by drastically reducing its gradient estimator bias","volume":"15","author":"Laborieux","year":"2021","journal-title":"Front. Neurosci."},{"key":"nceadebaabib38","article-title":"Updates of equilibrium prop match gradients of backprop through time in an RNN with static input","volume":"vol 32","author":"Ernoult","year":"2019"},{"key":"nceadebaabib39","article-title":"An overview of gradient descent optimization algorithms","author":"Ruder","year":"2017"},{"key":"nceadebaabib40","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2014"},{"key":"nceadebaabib41","doi-asserted-by":"publisher","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"LeCun","year":"1998","journal-title":"Proc. IEEE"},{"key":"nceadebaabib42","doi-asserted-by":"publisher","first-page":"576","DOI":"10.1143\/PTP.76.576","article-title":"A soluble active rotater model showing phase transitions via mutual entertainment","volume":"76","author":"Sakaguchi","year":"1986","journal-title":"Prog. Theor. Phys."},{"key":"nceadebaabib43","doi-asserted-by":"publisher","DOI":"10.1063\/5.0041575","article-title":"Ultrafast Ising machines using spin torque nano-oscillators","volume":"118","author":"Albertsson","year":"2021","journal-title":"Appl. Phys. Lett."}],"container-title":["Neuromorphic Computing and Engineering"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/adebaa","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/adebaa\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/adebaa","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/adebaa\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/adebaa\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/adebaa\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/adebaa\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/adebaa\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,25]],"date-time":"2025-07-25T10:03:03Z","timestamp":1753437783000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/adebaa"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,25]]},"references-count":43,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2025,7,25]]},"published-print":{"date-parts":[[2025,9,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2634-4386\/adebaa","relation":{},"ISSN":["2634-4386"],"issn-type":[{"value":"2634-4386","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,25]]},"assertion":[{"value":"Training and synchronizing oscillator networks with Equilibrium Propagation","name":"article_title","label":"Article Title"},{"value":"Neuromorphic Computing and Engineering","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2025 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2025-04-16","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-07-03","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-07-25","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}