Digital-analog hybrid matrix multiplication processor for optical neural networks

Digital-analog hybrid matrix multiplication processor for optical neural networks | 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 Digital-analog hybrid matrix multiplication processor for optical neural networks Deming Kong, Xiansong Meng, Kwangwoong Kim, Qiuchi Li, Po Dong, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4003496/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract The computational demands of modern AI have spurred interest in optical neural networks (ONNs) which offers the potential benefits of increased speed and lower power consumption. However, current ONNs running analog optical signals face a fundamental limitation in calculation precision, typically around 4 bits, due to accumulated noise from electro-optical components. This obstacle is inherent to the analog computing nature of ONNs. The analog signals representing the inputs and the computational results require high-resolution signal format converters (digital-to-analogue conversions (DACs) and analogue-to-digital conversions (ADCs)), which pose a major obstacle in practical implementation. Here, we propose a digital analog hybrid optical computing architecture for ONNs, which utilizes digital optical inputs in the form of binary words. By introducing the logic levels and decisions based on thresholding, the calculation precision is significantly enhanced. The DACs for inputs are removed. And the resolution requirement of the ADCs becomes independent of the bitwidth of the inputs. This can increase the operating speed at a high calculation precision and facilitate compatibility with microelectronics. To validate our approach, we have fabricated a proof-of-concept photonic chip and built up a hybrid optical processor (HOP) system for neural network applications. We have demonstrated an unprecedented 16-bit calculation precision for high-definition image processing, with a pixel error rate (PER) as low as 1.8×10 −3 at a signal-to-noise ratio (SNR) of 18.2 dB. We have also implemented a convolutional neural network for handwritten digit recognition that shows the same accuracy as the one achieved by a desktop computer. The concept of the digital-analog hybrid optical computing architecture offers a methodology that could potentially be applied to various ONN implementations and may intrigue new research into efficient and accurate domain-specific optical computing architectures for neural networks. Physical sciences/Physics/Electronics, photonics and device physics/Photonic devices Physical sciences/Optics and photonics/Applied optics/Integrated optics Physical sciences/Optics and photonics/Optical materials and structures/Silicon photonics Full Text Additional Declarations There is NO Competing Interest. Cite Share Download PDF Status: Under Review 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-4003496","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":279356586,"identity":"4b049151-331a-4ce2-a351-980ecebad9f7","order_by":0,"name":"Deming Kong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAkklEQVRIiWNgGAWjYDCCA2xAooIZxDQmRcsZZgkStTC2kaKF79qxxMeF86zrDA4wbzYgSovk7bTDxjO3pUsYHGArTiBKi8Ht9DZp3m2HgVp4jA+QoGUOaVrSjknzNkC0EOcwoF+SjXmOpUvOPMxWTJz3+W6nGT7mqbHm5zvevFmCKC0IwEyi+lEwCkbBKBgFeAAAIqAsdEZl8p0AAAAASUVORK5CYII=","orcid":"","institution":"Technical University of Denmark","correspondingAuthor":true,"prefix":"","firstName":"Deming","middleName":"","lastName":"Kong","suffix":""},{"id":279356587,"identity":"9c59bf67-fec6-48b0-82f0-d95f160cf40c","order_by":1,"name":"Xiansong Meng","email":"","orcid":"","institution":"Technical University of Denmark","correspondingAuthor":false,"prefix":"","firstName":"Xiansong","middleName":"","lastName":"Meng","suffix":""},{"id":279356588,"identity":"f8d648e8-2287-45cd-b83d-d8fba501e60d","order_by":2,"name":"Kwangwoong Kim","email":"","orcid":"","institution":"Nokia (United States)","correspondingAuthor":false,"prefix":"","firstName":"Kwangwoong","middleName":"","lastName":"Kim","suffix":""},{"id":279356589,"identity":"c7c003df-5cf7-4f9c-92b9-db3d79726ac5","order_by":3,"name":"Qiuchi Li","email":"","orcid":"","institution":"University of Copenhagen","correspondingAuthor":false,"prefix":"","firstName":"Qiuchi","middleName":"","lastName":"Li","suffix":""},{"id":279356590,"identity":"b6e174c2-7abf-4de5-84d8-359a1e32af5d","order_by":4,"name":"Po Dong","email":"","orcid":"","institution":"II-VI Incorporated","correspondingAuthor":false,"prefix":"","firstName":"Po","middleName":"","lastName":"Dong","suffix":""},{"id":279356591,"identity":"e7760060-d51d-415e-9ae9-5ff530bae0a1","order_by":5,"name":"Ingemar Cox","email":"","orcid":"","institution":"University of Copenhagen","correspondingAuthor":false,"prefix":"","firstName":"Ingemar","middleName":"","lastName":"Cox","suffix":""},{"id":279356592,"identity":"1bc63736-548e-400b-9d3d-370ef7db4d20","order_by":6,"name":"Christina Lioma","email":"","orcid":"","institution":"University of Copenhagen","correspondingAuthor":false,"prefix":"","firstName":"Christina","middleName":"","lastName":"Lioma","suffix":""},{"id":279356593,"identity":"3cc214f6-a524-4455-b5e1-c24cfc5aa44d","order_by":7,"name":"Hao Hu","email":"","orcid":"","institution":"Technical University of Denmark","correspondingAuthor":false,"prefix":"","firstName":"Hao","middleName":"","lastName":"Hu","suffix":""}],"badges":[],"createdAt":"2024-03-01 14:31:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4003496/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4003496/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":52952947,"identity":"01d82097-8fcd-4068-995e-0c5dadbb6a6c","added_by":"auto","created_at":"2024-03-19 04:31:02","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5580884,"visible":true,"origin":"","legend":"","description":"","filename":"02mergedmaintextwithsupplementaryinfo.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4003496/v1_covered_31071ec2-580d-469a-9de4-b257d2d0f5c1.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Digital-analog hybrid matrix multiplication processor for optical neural networks","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-4003496/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4003496/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The computational demands of modern AI have spurred interest in optical neural networks (ONNs) which offers the potential benefits of increased speed and lower power consumption. However, current ONNs running analog optical signals face a fundamental limitation in calculation precision, typically around 4 bits, due to accumulated noise from electro-optical components. This obstacle is inherent to the analog computing nature of ONNs. The analog signals representing the inputs and the computational results require high-resolution signal format converters (digital-to-analogue conversions (DACs) and analogue-to-digital conversions (ADCs)), which pose a major obstacle in practical implementation. Here, we propose a digital analog hybrid optical computing architecture for ONNs, which utilizes digital optical inputs in the form of binary words. By introducing the logic levels and decisions based on thresholding, the calculation precision is significantly enhanced. The DACs for inputs are removed. And the resolution requirement of the ADCs becomes independent of the bitwidth of the inputs. This can increase the operating speed at a high calculation precision and facilitate compatibility with microelectronics. To validate our approach, we have fabricated a proof-of-concept photonic chip and built up a hybrid optical processor (HOP) system for neural network applications. We have demonstrated an unprecedented 16-bit calculation precision for high-definition image processing, with a pixel error rate (PER) as low as 1.8×10\u003csup\u003e−3\u003c/sup\u003e at a signal-to-noise ratio (SNR) of 18.2 dB. We have also implemented a convolutional neural network for handwritten digit recognition that shows the same accuracy as the one achieved by a desktop computer. The concept of the digital-analog hybrid optical computing architecture offers a methodology that could potentially be applied to various ONN implementations and may intrigue new research into efficient and accurate domain-specific optical computing architectures for neural networks.","manuscriptTitle":"Digital-analog hybrid matrix multiplication processor for optical neural networks","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-03-19 04:22:47","doi":"10.21203/rs.3.rs-4003496/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":"3c073aca-6827-436e-8a9e-0ee60632f703","owner":[],"postedDate":"March 19th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":29421101,"name":"Physical sciences/Physics/Electronics, photonics and device physics/Photonic devices"},{"id":29421102,"name":"Physical sciences/Optics and photonics/Applied optics/Integrated optics"},{"id":29421103,"name":"Physical sciences/Optics and photonics/Optical materials and structures/Silicon photonics"}],"tags":[],"updatedAt":"2024-03-19T04:22:48+00:00","versionOfRecord":[],"versionCreatedAt":"2024-03-19 04:22:47","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4003496","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4003496","identity":"rs-4003496","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}