Brain-inspired synaptic transistors for in-situ spiking reinforcement learning with eligibility trace

Brain-inspired synaptic transistors for in-situ spiking reinforcement learning with eligibility trace | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Brain-inspired synaptic transistors for in-situ spiking reinforcement learning with eligibility trace Yuhui He, Yasai Wang, Weiwei Xiong, Jianmin Yan, Yue Zhou, Chaoyi Zhu, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6296374/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 21 Feb, 2026 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Abstract Brain-inspired reinforcement learning (RL) represents a pivotal pathway toward artificial general intelligence, yet existing hardware implementations based on artificial neural networks lack critical biological mechanisms like third-terminal modulated eligibility traces and dynamic reward signaling. Emerging materials can address these challenges by mimicking RL’s complex dynamics with revolutionary efficiency. Here we demonstrate a brain-inspired SNN-based RL computing architecture using α-In 2 Se 3 ferroelectric semiconductor field-effect transistor (FeS-FET). By leveraging the intrinsic in-plane and out-of-plane polarization coupling of α-In 2 Se 3 , the multi-terminal conductance modulation in the FeS-FET enables reward signal modulation of RL. The ferroelectric relaxation is utilized to implement biological eligibility trace decay, thereby enhancing the algorithm's processing capability. autonomous driving tasks are then demonstrated with RL neural network constructed by the α-In 2 Se 3 FeS-FET array, where in-situ reward-based weight updates and eligibility trace decay are performed without any external memory or computing units. Our solution paves the way for a SNN-based RL computing architecture with full functionality, low energy consumption and reduced hardware overhead. Physical sciences/Nanoscience and technology/Nanoscale materials/Two-dimensional materials Physical sciences/Materials science/Materials for devices/Information storage Full Text Additional Declarations There is NO Competing Interest. Supplementary Files InSeSupplementaryInformation.docx Supplementary Information Cite Share Download PDF Status: Published Journal Publication published 21 Feb, 2026 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6296374","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":435280195,"identity":"62444057-9e9f-4076-bebc-590cdc634e59","order_by":0,"name":"Yuhui He","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2ElEQVRIiWNgGAWjYBACAyBmZmCQkDOA8JmJ12IMZDA2kKKFIXED0VrM2XvMpAvbLNK3S6Q/f8BQYZ3YwH72AF4tlj1nzKRntknk7pyRY9jAcCY9sYEnLwG/w27kmEnzArVsuJHD2MDYdjixQYLHAL+W+2/AWtINbqQ/bGD8R4yWGzxgLQkGNxIMGxgbiNFyJq3YmuechOGGM28MZyQcSzdu48khoOX44Y23ecrq5A2Opz/48KHGWraf/Qx+LQwMHEgKEoCYjYB6IGB/QFjNKBgFo2AUjGwAAJCMQwcxlJquAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-0546-7692","institution":"Huazhong University of Science and Technology,","correspondingAuthor":true,"prefix":"","firstName":"Yuhui","middleName":"","lastName":"He","suffix":""},{"id":435280196,"identity":"94b0441c-6736-4803-a52e-8d24bca3a1e6","order_by":1,"name":"Yasai Wang","email":"","orcid":"","institution":"Huazhong University of Science and Technology,The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Yasai","middleName":"","lastName":"Wang","suffix":""},{"id":435280197,"identity":"b18b3957-508a-4276-8ffa-84901562942b","order_by":2,"name":"Weiwei Xiong","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Weiwei","middleName":"","lastName":"Xiong","suffix":""},{"id":435280198,"identity":"87f602f2-7e64-4274-8715-4e9f4593f84b","order_by":3,"name":"Jianmin Yan","email":"","orcid":"https://orcid.org/0000-0002-2467-630X","institution":"Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Jianmin","middleName":"","lastName":"Yan","suffix":""},{"id":435280199,"identity":"5f4c94b4-d2ee-4fe2-a9b5-fb76d8559202","order_by":4,"name":"Yue Zhou","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Yue","middleName":"","lastName":"Zhou","suffix":""},{"id":435280200,"identity":"0f00455d-9cbf-4e6c-996a-2b5e8be7bf92","order_by":5,"name":"Chaoyi Zhu","email":"","orcid":"https://orcid.org/0000-0001-5119-5512","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Chaoyi","middleName":"","lastName":"Zhu","suffix":""},{"id":435280201,"identity":"0a981932-8f38-4de8-8996-f64534b791bb","order_by":6,"name":"Xiangshui Miao","email":"","orcid":"https://orcid.org/0000-0002-3999-7421","institution":"Huazhong University of Science and Technology/Hubei Yangtze Memory Laboratories","correspondingAuthor":false,"prefix":"","firstName":"Xiangshui","middleName":"","lastName":"Miao","suffix":""},{"id":435280202,"identity":"81010942-bf9e-48c1-b04d-7b7b232cabea","order_by":7,"name":"Yang Chai","email":"","orcid":"https://orcid.org/0000-0002-8943-0861","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Chai","suffix":""}],"badges":[],"createdAt":"2025-03-24 14:37:36","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6296374/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6296374/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41467-026-69898-9","type":"published","date":"2026-02-21T05:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":105789210,"identity":"bfd9f855-4558-4e91-ab18-94b819bb4517","added_by":"auto","created_at":"2026-03-31 07:13:33","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1200298,"visible":true,"origin":"","legend":"","description":"","filename":"Braininspiredsynaptictransistorsforinsituspikingreinforcementlearningwitheligibilitytrace.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6296374/v1_covered_5f30eea3-f2a6-44ae-a223-7c11315c7538.pdf"},{"id":79537495,"identity":"672d938f-4831-417e-b830-9164baa59df3","added_by":"auto","created_at":"2025-03-31 02:19:59","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":6415514,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"InSeSupplementaryInformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-6296374/v1/345458c0f23137e4ac2e03ad.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Brain-inspired synaptic transistors for in-situ spiking reinforcement learning with eligibility trace","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6296374/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6296374/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eBrain-inspired reinforcement learning (RL) represents a pivotal pathway toward artificial general intelligence, yet existing hardware implementations based on artificial neural networks lack critical biological mechanisms like third-terminal modulated eligibility traces and dynamic reward signaling. Emerging materials can address these challenges by mimicking RL’s complex dynamics with revolutionary efficiency. Here we demonstrate a brain-inspired SNN-based RL computing architecture using α-In\u003csub\u003e2\u003c/sub\u003eSe\u003csub\u003e3\u003c/sub\u003e ferroelectric semiconductor field-effect transistor (FeS-FET). By leveraging the intrinsic in-plane and out-of-plane polarization coupling of α-In\u003csub\u003e2\u003c/sub\u003eSe\u003csub\u003e3\u003c/sub\u003e, the multi-terminal conductance modulation in the FeS-FET enables reward signal modulation of RL. The ferroelectric relaxation is utilized to implement biological eligibility trace decay, thereby enhancing the algorithm's processing capability. autonomous driving tasks are then demonstrated with RL neural network constructed by the α-In\u003csub\u003e2\u003c/sub\u003eSe\u003csub\u003e3\u003c/sub\u003e FeS-FET array, where in-situ reward-based weight updates and eligibility trace decay are performed without any external memory or computing units. Our solution paves the way for a SNN-based RL computing architecture with full functionality, low energy consumption and reduced hardware overhead.\u003c/p\u003e","manuscriptTitle":"Brain-inspired synaptic transistors for in-situ spiking reinforcement learning with eligibility trace","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-31 02:11:54","doi":"10.21203/rs.3.rs-6296374/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"c486c985-e486-42d9-85f4-164368543c27","owner":[],"postedDate":"March 31st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":46349702,"name":"Physical sciences/Nanoscience and technology/Nanoscale materials/Two-dimensional materials"},{"id":46349703,"name":"Physical sciences/Materials science/Materials for devices/Information storage"}],"tags":[],"updatedAt":"2026-03-31T07:13:02+00:00","versionOfRecord":{"articleIdentity":"rs-6296374","link":"https://doi.org/10.1038/s41467-026-69898-9","journal":{"identity":"nature-communications","isVorOnly":false,"title":"Nature Communications"},"publishedOn":"2026-02-21 05:00:00","publishedOnDateReadable":"February 21st, 2026"},"versionCreatedAt":"2025-03-31 02:11:54","video":"","vorDoi":"10.1038/s41467-026-69898-9","vorDoiUrl":"https://doi.org/10.1038/s41467-026-69898-9","workflowStages":[]},"version":"v1","identity":"rs-6296374","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6296374","identity":"rs-6296374","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}