A Kolmogorov–Arnold Compute-in-Memory (KA-CIM) Hardware Accelerator with High Energy Efficiency and Flexibility

A Kolmogorov–Arnold Compute-in-Memory (KA-CIM) Hardware Accelerator with High Energy Efficiency and Flexibility | 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 A Kolmogorov–Arnold Compute-in-Memory (KA-CIM) Hardware Accelerator with High Energy Efficiency and Flexibility Chirag Sudarshan, Paul Manea, John Paul Strachan This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5804189/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Kolmogorov-Arnold Networks (KAN) are an emerging AI model designed for AI+Science applications, offering up to 100x fewer parameters than conventional Multilayer Perceptrons (MLPs). KAN relies on computationally expensive non-linear functions, unlike MLPs, which are dominated by matrix multiplication. This limits KAN's compatibility with energy-efficient hardware accelerators like Compute-in-Memory (CIM), restricting it to energy-inefficient general-purpose chips. To address this gap, we propose KA-CIM, a memory-centric design for energy-efficient computation of KAN inference. We leverage the Piece-Wise Linear (PWL) approximation of non-linear functions to convert complex computations into Multiply-Accumulate (MAC) operations and pre-storing segment parameters, an approach suitable for memory-centric design. A specialized CIM unit dynamically routes inputs to appropriate PWL segments, while a crossbar array retrieves segment slope and intercept to execute the MAC operations. The tile partitioning feature facilitates higher PWL segments for improved accuracy without significant energy penalties. This architectural design facilitates efficient and flexible computation of arbitrary non-linear functions of KANs. Beyond KAN inference, KA-CIM's capability extends to multi-variable equations and derivative computations. KA-CIM achieves energy-delay products 1073x lower than CPU for non-linear operations. When executing KAN, it achieves 77x lower energy-delay product than a 100 TOPS/W CIM accelerator executing MLP for the same task. Scientific community and society/Scientific community/Publishing/Authorship Scientific community and society/Scientific community/Publishing/Peer review Full Text Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryMaterial.pdf A Kolmogorov–Arnold Compute-in-Memory (KA-CIM) Hardware Accelerator with High Energy Efficiency and Flexibility Cite Share Download PDF Status: Posted 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-5804189","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":403044816,"identity":"dd5d07e1-746d-4ad6-80c4-338b5cc2951b","order_by":0,"name":"Chirag Sudarshan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABFklEQVRIiWNgGAWjYLACHiA2YGBgfFDBwMAMRA0HGBsYgIiwFmaDM2AtjMRrYZM4A+ZC1OPUwt/A/kziTc09e3P2s8cqDvxhYDc43th44OcOBtl+HFokDvCYSc45Vpy4sycv7cYBHpDrDjYc7D3DYDwThzUGDDxs0jxsCQkGB3LMbn+Q+M9scCOx4TBjG0PihgO4tLA/k+b5l2BvcP6NWcEBA6At9x9CtOzHqYXBTJq3LYFxw40cM4YDCUAtNxihtuDyy2EeY8u5fQmJG268MZY4cICBWfJMItAvbRLGM3DYwt/e/vDGm28gh+UYfgCGWDLf8cOHP/xss5Htx+F9UNShgGSY9TjUYwF2xCsdBaNgFIyCkQIALLFhDFdlN0cAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-1651-1935","institution":"Forschungszentrum Jülich GmbH","correspondingAuthor":true,"prefix":"","firstName":"Chirag","middleName":"","lastName":"Sudarshan","suffix":""},{"id":403044817,"identity":"6aa7c5a7-b168-4660-bbc2-dcd9209f72a6","order_by":1,"name":"Paul Manea","email":"","orcid":"","institution":"Forschungszentrum Jülich GmbH","correspondingAuthor":false,"prefix":"","firstName":"Paul","middleName":"","lastName":"Manea","suffix":""},{"id":403044818,"identity":"07528943-a81c-40aa-a05e-c0052b0adbb6","order_by":2,"name":"John Paul Strachan","email":"","orcid":"https://orcid.org/0000-0002-1382-3677","institution":"Forschungszentrum Jülich GmbH","correspondingAuthor":false,"prefix":"","firstName":"John","middleName":"Paul","lastName":"Strachan","suffix":""}],"badges":[],"createdAt":"2025-01-10 13:50:46","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5804189/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5804189/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":83982028,"identity":"5faac8f4-f6bf-42b8-9f60-91ab4f850d04","added_by":"auto","created_at":"2025-06-05 10:19:13","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":19025425,"visible":true,"origin":"","legend":"Article File","description":"","filename":"Paper.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5804189/v1_covered_51ebff20-9345-4976-b003-654a50d9dafe.pdf"},{"id":74014357,"identity":"8a1b8ec5-5987-4fbe-9114-c000434f6ac2","added_by":"auto","created_at":"2025-01-17 03:26:53","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":4713134,"visible":true,"origin":"","legend":"A Kolmogorov\u0026#x2013;Arnold Compute-in-Memory (KA-CIM) Hardware Accelerator with High Energy Efficiency and Flexibility","description":"","filename":"SupplementaryMaterial.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5804189/v1/aa980b6ea3dc6f3ba5498df5.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"A Kolmogorov–Arnold Compute-in-Memory (KA-CIM) Hardware Accelerator with High Energy Efficiency and Flexibility","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5804189/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5804189/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Kolmogorov-Arnold Networks (KAN) are an emerging AI model designed for AI+Science applications, offering up to 100x fewer parameters than conventional Multilayer Perceptrons (MLPs). KAN relies on computationally expensive non-linear functions, unlike MLPs, which are dominated by matrix multiplication. This limits KAN's compatibility with energy-efficient hardware accelerators like Compute-in-Memory (CIM), restricting it to energy-inefficient general-purpose chips. To address this gap, we propose KA-CIM, a memory-centric design for energy-efficient computation of KAN inference. We leverage the Piece-Wise Linear (PWL) approximation of non-linear functions to convert complex computations into Multiply-Accumulate (MAC) operations and pre-storing segment parameters, an approach suitable for memory-centric design. A specialized CIM unit dynamically routes inputs to appropriate PWL segments, while a crossbar array retrieves segment slope and intercept to execute the MAC operations. The tile partitioning feature facilitates higher PWL segments for improved accuracy without significant energy penalties. This architectural design facilitates efficient and flexible computation of arbitrary non-linear functions of KANs. Beyond KAN inference, KA-CIM's capability extends to multi-variable equations and derivative computations. KA-CIM achieves energy-delay products 1073x lower than CPU for non-linear operations. When executing KAN, it achieves 77x lower energy-delay product than a 100 TOPS/W CIM accelerator executing MLP for the same task.","manuscriptTitle":"A Kolmogorov–Arnold Compute-in-Memory (KA-CIM) Hardware Accelerator with High Energy Efficiency and Flexibility","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-01-17 03:26:49","doi":"10.21203/rs.3.rs-5804189/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"0faf6756-057a-493b-b963-d7815026e658","owner":[],"postedDate":"January 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":42969656,"name":"Scientific community and society/Scientific community/Publishing/Authorship"},{"id":42969657,"name":"Scientific community and society/Scientific community/Publishing/Peer review"}],"tags":[],"updatedAt":"2025-06-05T10:10:52+00:00","versionOfRecord":[],"versionCreatedAt":"2025-01-17 03:26:49","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5804189","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5804189","identity":"rs-5804189","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}