{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,5,8]],"date-time":"2025-05-08T12:10:08Z","timestamp":1746706208901,"version":"3.40.5"},"reference-count":52,"publisher":"IOP Publishing","issue":"2","license":[{"start":{"date-parts":[[2025,5,8]],"date-time":"2025-05-08T00:00:00Z","timestamp":1746662400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,5,8]],"date-time":"2025-05-08T00:00:00Z","timestamp":1746662400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/501100000038","name":"Natural Sciences and Engineering Research Council of Canada","doi-asserted-by":"crossref","award":["559730"],"award-info":[{"award-number":["559730"]}],"id":[{"id":"10.13039\/501100000038","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100021202","name":"Alliance de recherche num\u00e9rique du Canada","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100021202","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100003151","name":"Fonds de recherche du Qu\u00e9bec \u2013 Nature et technologies","doi-asserted-by":"crossref","award":["287330"],"award-info":[{"award-number":["287330"]}],"id":[{"id":"10.13039\/501100003151","id-type":"DOI","asserted-by":"crossref"}]},{"name":"IONOS-ERC","award":["773228"],"award-info":[{"award-number":["773228"]}]},{"name":"Minist\u00e8re de l\u2019\u00c9conomie, de l\u2019Innovation et de l\u2019\u00c9nergie"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Neuromorph. Comput. Eng."],"published-print":{"date-parts":[[2025,6,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Training Recurrent Spiking Neural Networks (RSNNs) with binary spikes for tasks of extended time scales presents a challenge due to the amplified vanishing gradient problem during back propagation through time. This paper introduces three crucial elements that significantly enhance the memory and capabilities of RSNNs, with a strong emphasis on compatibility with hardware and neuromorphic systems. Firstly, we incorporate neuron-level synaptic delays, which not only allow the gradient to skip time steps but also reduce the overall neuron population\u2019s firing rate. Subsequently, we apply a biologically inspired branching factor regularization rule to stabilize the network\u2019s dynamics and make training easier by incorporating a time-local error in the loss function. Lastly, we modify a commonly used surrogate gradient function by increasing its support to facilitate learning over longer timescales when using binary spikes. By integrating these three innovative elements, we not only resolve several complex benchmarks but also achieve state-of-the-art results on the spiking permuted sequential MNIST task (psMNIST), showcasing the practicality and relevance of our approach for digital and analog neuromorphic systems.<\/jats:p>","DOI":"10.1088\/2634-4386\/add293","type":"journal-article","created":{"date-parts":[[2025,5,1]],"date-time":"2025-05-01T08:13:13Z","timestamp":1746087193000},"page":"024008","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Enhancing temporal learning in recurrent spiking networks for neuromorphic applications"],"prefix":"10.1088","volume":"5","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6730-0794","authenticated-orcid":true,"given":"Ismael","family":"Balafrej","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0216-0457","authenticated-orcid":true,"given":"Soufiyan","family":"Bahadi","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9306-426X","authenticated-orcid":false,"given":"Jean","family":"Rouat","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9591-220X","authenticated-orcid":false,"given":"Fabien","family":"Alibart","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"266","published-online":{"date-parts":[[2025,5,8]]},"reference":[{"year":"1991","author":"Abeles","key":"nceadd293bib1","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511574566"},{"key":"nceadd293bib2","doi-asserted-by":"publisher","DOI":"10.1088\/2634-4386\/ac6533","article-title":"P-critical: A reservoir autoregulation plasticity rule for neuromorphic hardware","volume":"2","author":"Balafrej","year":"2022","journal-title":"Neuromorph. Comput. Eng."},{"key":"nceadd293bib3","doi-asserted-by":"publisher","first-page":"329","DOI":"10.1098\/rsta.2007.2092","article-title":"The criticality hypothesis: how local cortical networks might optimize information processing","volume":"366","author":"Beggs","year":"2007","journal-title":"Phil. Trans. R. Soc. A"},{"article-title":"Long short-term memory and learning-to-learn in networks of spiking neurons","year":"2018","author":"Bellec","key":"nceadd293bib4"},{"key":"nceadd293bib5","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41467-020-17236-y","article-title":"A solution to the learning dilemma for recurrent networks of spiking neurons","volume":"11","author":"Bellec","year":"2020","journal-title":"Nat. Commun."},{"key":"nceadd293bib6","doi-asserted-by":"publisher","first-page":"157","DOI":"10.1109\/72.279181","article-title":"Learning long-term dependencies with gradient descent is difficult","volume":"5","author":"Bengio","year":"1994","journal-title":"IEEE Trans. Neural Netw."},{"key":"nceadd293bib7","doi-asserted-by":"crossref","DOI":"10.1145\/3320288.3320304","article-title":"Benchmarking keyword spotting efficiency on neuromorphic hardware","author":"Blouw","year":"2019"},{"key":"nceadd293bib8","doi-asserted-by":"publisher","first-page":"1926","DOI":"10.1126\/science.1099745","article-title":"Neuronal oscillations in cortical networks","volume":"304","author":"Buzs\u00e1ki","year":"2004","journal-title":"Science"},{"key":"nceadd293bib9","first-page":"1898","article-title":"Parallelizing Legendre Memory Unit Training","author":"Chilkuri","year":"2021"},{"key":"nceadd293bib10","doi-asserted-by":"publisher","first-page":"3446","DOI":"10.1038\/s41467-024-47764-w","article-title":"DenRAM: neuromorphic dendritic architecture with RRAM for efficient temporal processing with delays","volume":"15","author":"D\u2019Agostino","year":"2024","journal-title":"Nat. Commun."},{"key":"nceadd293bib11","doi-asserted-by":"publisher","DOI":"10.1063\/1.5108663","article-title":"Hybrid neuromorphic circuits exploiting non-conventional properties of RRAM for massively parallel local plasticity mechanisms","volume":"7","author":"Dalgaty","year":"2019","journal-title":"APL Mater."},{"key":"nceadd293bib12","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1109\/MM.2018.112130359","article-title":"Loihi: a neuromorphic manycore processor with on-chip learning","volume":"38","author":"Davies","year":"2018","journal-title":"IEEE Micro"},{"key":"nceadd293bib13","doi-asserted-by":"publisher","first-page":"911","DOI":"10.1109\/JPROC.2021.3067593","article-title":"Advancing neuromorphic computing with loihi: a survey of results and outlook","volume":"109","author":"Davies","year":"2021","journal-title":"Proc. IEEE"},{"key":"nceadd293bib14","first-page":"1","article-title":"A momentumized, adaptive, dual averaged gradient method","volume":"23","author":"Defazio","year":"2022","journal-title":"J. Mach. Learn. Res."},{"key":"nceadd293bib15","doi-asserted-by":"crossref","DOI":"10.1109\/ISCAS51556.2021.9401446","article-title":"PCM-trace: scalable synaptic eligibility traces with resistivity drift of phase-change materials","author":"Demirag","year":"2021"},{"key":"nceadd293bib16","doi-asserted-by":"publisher","first-page":"509","DOI":"10.1038\/s41586-019-1261-9","article-title":"Specialized coding of sensory, motor and cognitive variables in VTA dopamine neurons","volume":"570","author":"Engelhard","year":"2019","journal-title":"Nature"},{"article-title":"PyTorch e-prop implementation","year":"2022","author":"Frenkel","key":"nceadd293bib17"},{"key":"nceadd293bib18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/ISSCC42614.2022.9731734","article-title":"ReckOn: a 28nm Sub-mm2 Task-Agnostic spiking recurrent neural network processor enabling on-chip learning over second-long timescales","volume":"vol 65","author":"Frenkel","year":"2022"},{"key":"nceadd293bib19","doi-asserted-by":"publisher","first-page":"2251","DOI":"10.1162\/NECO_a_00331","article-title":"Frequency selectivity emerging from spike-timing-dependent plasticity","volume":"24","author":"Gilson","year":"2012","journal-title":"Neural Comput."},{"year":"2016","author":"Goodfellow","key":"nceadd293bib20"},{"key":"nceadd293bib21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1103\/PhysRevLett.94.058101","article-title":"Critical branching captures activity in living neural networks and maximizes the number of metastable states","volume":"94","author":"Haldeman","year":"2005","journal-title":"Phys. Rev. Lett."},{"key":"nceadd293bib22","doi-asserted-by":"publisher","DOI":"10.1002\/adfm.202204102","article-title":"A review of artificial spiking neuron devices for neural processing and sensing","volume":"32","author":"Han","year":"2022","journal-title":"Adv. Funct. Mater."},{"article-title":"Deep residual learning for image recognition","year":"2015a","author":"He","key":"nceadd293bib23"},{"key":"nceadd293bib24","doi-asserted-by":"crossref","DOI":"10.1109\/ICCV.2015.123","article-title":"Delving deep into rectifiers: surpassing human-level performance on imagenet classification","author":"He","year":"2015b"},{"article-title":"Stabilizing spiking neuron training","year":"2022","author":"Herranz-Celotti","key":"nceadd293bib25"},{"key":"nceadd293bib26","first-page":"p 34","article-title":"Increasing liquid state machine performance with edge-of-chaos dynamics organized by astrocyte-modulated plasticity","author":"Ivanov","year":"2021"},{"key":"nceadd293bib27","doi-asserted-by":"publisher","first-page":"245","DOI":"10.1162\/089976606775093882","article-title":"Polychronization: computation with spikes","volume":"18","author":"Izhikevich","year":"2006","journal-title":"Neural Comput."},{"key":"nceadd293bib28","first-page":"55","article-title":"SSO-LSM: a sparse and self-organizing architecture for liquid state machine based neural processors","volume":"vol 2016","author":"Jin","year":"2016"},{"key":"nceadd293bib29","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR52729.2023.00992","article-title":"SMPConv: self-moving point representations for continuous convolution","author":"Kim","year":"2023"},{"article-title":"A simple way to initialize recurrent networks of rectified linear units","year":"2015","author":"Le","key":"nceadd293bib30"},{"key":"nceadd293bib31","doi-asserted-by":"publisher","first-page":"10141","DOI":"10.1109\/TNNLS.2022.3164930","article-title":"Supervised learning in multilayer spiking neural networks with spike temporal error backpropagation","volume":"34","author":"Luo","year":"2023","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"nceadd293bib32","doi-asserted-by":"publisher","first-page":"2531","DOI":"10.1162\/089976602760407955","article-title":"Real-time computing without stable states: a new framework for neural computation based on perturbations","volume":"14","author":"Maass","year":"2002","journal-title":"Neural Comput."},{"key":"nceadd293bib33","doi-asserted-by":"publisher","first-page":"1672","DOI":"10.1038\/nn.4403","article-title":"History-dependent variability in population dynamics during evidence accumulation in cortex","volume":"19","author":"Morcos","year":"2016","journal-title":"Nat. Neurosci."},{"key":"nceadd293bib34","doi-asserted-by":"publisher","first-page":"51","DOI":"10.1109\/MSP.2019.2931595","article-title":"Surrogate gradient learning in spiking neural networks: bringing the power of gradient-based optimization to spiking neural networks","volume":"36","author":"Neftci","year":"2019","journal-title":"IEEE Signal Process. Mag."},{"key":"nceadd293bib35","doi-asserted-by":"publisher","first-page":"437","DOI":"10.3389\/fnins.2015.00437","article-title":"Converting static image datasets to spiking neuromorphic datasets using saccades","volume":"9","author":"Orchard","year":"2015","journal-title":"Front. Neurosci."},{"key":"nceadd293bib36","first-page":"1310","article-title":"On the difficulty of training recurrent neural networks","author":"Pascanu","year":"2013"},{"key":"nceadd293bib37","article-title":"PyTorch: an imperative style, high-performance deep learning library","volume":"vol 32","author":"Paszke","year":"2019"},{"article-title":"Hardware-aware training of models with synaptic delays for digital event-driven neuromorphic processors","year":"2024","author":"Patino-Saucedo","key":"nceadd293bib38"},{"key":"nceadd293bib39","doi-asserted-by":"publisher","first-page":"5793","DOI":"10.1038\/s41467-022-33476-6","article-title":"Self-organization of an inhomogeneous memristive hardware for sequence learning","volume":"13","author":"Payvand","year":"2022","journal-title":"Nat. Commun."},{"key":"nceadd293bib40","doi-asserted-by":"publisher","first-page":"108","DOI":"10.3389\/fnsys.2014.00108","article-title":"Spike avalanches in vivo suggest a driven, slightly subcritical brain state","volume":"8","author":"Priesemann","year":"2014","journal-title":"Front. Syst. Neurosci."},{"key":"nceadd293bib41","doi-asserted-by":"publisher","DOI":"10.1088\/2634-4386\/ac97bb","article-title":"Fluctuation-driven initialization for spiking neural network training","volume":"2","author":"Rossbroich","year":"2022","journal-title":"Neuromorph. Comput. Eng."},{"key":"nceadd293bib42","doi-asserted-by":"publisher","first-page":"85","DOI":"10.1016\/j.neunet.2014.09.003","article-title":"Deep learning in neural networks: an overview","volume":"61","author":"Schmidhuber","year":"2015","journal-title":"Neural Netw."},{"key":"nceadd293bib43","doi-asserted-by":"publisher","first-page":"277","DOI":"10.1109\/IJCNN.2016.7727209","article-title":"Adaptive delay learning in SpikeProp based on delay convergence analysis","author":"Shrestha","year":"2016"},{"key":"nceadd293bib44","doi-asserted-by":"publisher","first-page":"1451","DOI":"10.4249\/scholarpedia.1451","article-title":"Axonal conduction delays","volume":"7","author":"Swadlow","year":"2012","journal-title":"Scholarpedia J."},{"key":"nceadd293bib45","doi-asserted-by":"publisher","first-page":"75","DOI":"10.1016\/j.jphysparis.2005.12.002","article-title":"Glial cells in synaptic plasticity","volume":"99","author":"Todd","year":"2006","journal-title":"J. Physiol."},{"key":"nceadd293bib46","first-page":"p 32","article-title":"Legendre memory units: continuous-time representation in recurrent neural networks","author":"Voelker","year":"2019"},{"key":"nceadd293bib47","doi-asserted-by":"publisher","first-page":"328","DOI":"10.1109\/29.21701","article-title":"Phoneme recognition using time-delay neural networks","volume":"37","author":"Waibel","year":"1989","journal-title":"IEEE Trans. Acoust. Speech Signal Process."},{"key":"nceadd293bib48","doi-asserted-by":"publisher","first-page":"252","DOI":"10.3389\/fnins.2019.00252","article-title":"A Delay learning algorithm based on spike train kernels for spiking neurons","volume":"13","author":"Wang","year":"2019","journal-title":"Front. Neurosci."},{"key":"nceadd293bib49","doi-asserted-by":"publisher","first-page":"682","DOI":"10.1038\/s41583-020-00379-8","article-title":"Myelin plasticity: sculpting circuits in learning and memory","volume":"21","author":"Xin","year":"2020","journal-title":"Nat. Rev. Neurosci."},{"key":"nceadd293bib50","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3407197.3407225","article-title":"Effective and efficient computation with multiple-timescale spiking recurrent neural networks","author":"Yin","year":"2020"},{"key":"nceadd293bib51","doi-asserted-by":"publisher","first-page":"899","DOI":"10.1162\/neco_a_01367","article-title":"The remarkable robustness of surrogate gradient learning for instilling complex function in spiking neural networks","volume":"33","author":"Zenke","year":"2021","journal-title":"Neural Comput."},{"key":"nceadd293bib52","doi-asserted-by":"publisher","first-page":"103","DOI":"10.1016\/j.neucom.2020.03.079","article-title":"Supervised learning in spiking neural networks with synaptic delay-weight plasticity","volume":"409","author":"Zhang","year":"2020","journal-title":"Neurocomputing"}],"container-title":["Neuromorphic Computing and Engineering"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/add293","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/add293\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/add293","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/add293\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/add293\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/add293\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/add293\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/add293\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,5,8]],"date-time":"2025-05-08T11:51:42Z","timestamp":1746705102000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/add293"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,8]]},"references-count":52,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2025,5,8]]},"published-print":{"date-parts":[[2025,6,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2634-4386\/add293","relation":{},"ISSN":["2634-4386"],"issn-type":[{"type":"electronic","value":"2634-4386"}],"subject":[],"published":{"date-parts":[[2025,5,8]]},"assertion":[{"value":"Enhancing temporal learning in recurrent spiking networks for neuromorphic applications","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":"2025-01-11","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-04-30","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-05-08","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}