Rafale: Towards the Efficient Resource-Aware Federated Knowledge Distillation on Heterogeneous Clients

Rafale: Towards the Efficient Resource-Aware Federated Knowledge Distillation on Heterogeneous Clients | 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 Research Article Rafale: Towards the Efficient Resource-Aware Federated Knowledge Distillation on Heterogeneous Clients Harsh Pratap Singh, Debasis Das, Gozzal Eshniyazova, Doston Khasanov, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9548238/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 Knowledge distillation is an highly effective machine learning technique for transferring knowledge from a large model, i.e., teacher, to a smaller model, i.e., student, improving model efficiency without sacrificing performance. This paper proposes a novel resource-aware federated learning framework \textit{Rafale} that incorporates an adaptive client clustering mechanism based on computational resource availability and model complexity. High-resource clients train complex teacher models, and low-resource clients leverage distilled knowledge from teachers to train lightweight student models. Further, to optimize the aggregation process by introducing a clustering-based weight-sharing mechanism that minimizes communication overhead while maintaining model accuracy. In this paper, a revolutionary Rafale framework that greatly reduces communication cost and separates model training from architectural constraints is proposed. To increase scalability and robustness, the suggested method makes use of resource-aware client selection, adaptive information sharing, and effective distillation techniques. In comparison to state-of-the-art FL approaches, experimental data show that RA-FKD delivers greater accuracy, decreases communication costs by up to 76%, and improves system efficiency. A cosine similarity loss \textit{KDC} is introduced in knowledge distillation to improve the alignment between teacher and student models, enhancing knowledge transfer. Experimental evaluations on CIFAR-10 and CIFAR-100 datasets demonstrate that proposed framework achieving student test accuracy of 91.15% with reduction of 76.08% communication cost. Federated Learning Knowledge Distillation Resource-Aware Client Clustering Model Aggregation Communication Efficiency Full Text Additional Declarations No competing interests reported. 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-9548238","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":639210382,"identity":"46498134-de7c-4460-8eb0-1ee5086e8c1e","order_by":0,"name":"Harsh Pratap Singh","email":"","orcid":"","institution":"Indian Institute of Technology Jodhpur","correspondingAuthor":false,"prefix":"","firstName":"Harsh","middleName":"Pratap","lastName":"Singh","suffix":""},{"id":639210383,"identity":"67fecb3d-1358-4843-b367-834c316945b7","order_by":1,"name":"Debasis Das","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzElEQVRIiWNgGAWjYDCCAzxAwoBBjo0ZRAEBG7FajNmYmUnSwsCQ2MDATKS7+I73Hvz4o6AuvY+d/0ABQ40dA590A34tkmfOJUvzGBzObQM77FgyA5vMAfxaDG7kGEgzGByAamE7wMAmkUBAy/03xj9/GNSlQ7z/jxgtN3jMJHgMmBPAWhjbiNAieSYvzRroF0OgwwwMEvuSeQhq4Tt+9vDNH3/q5OX7Dz4z+PDNTk5+BgEtyIDNAKiYh3j1QMD8gCTlo2AUjIJRMGIAAGEHN1+REfH+AAAAAElFTkSuQmCC","orcid":"","institution":"Indian Institute of Technology Jodhpur","correspondingAuthor":true,"prefix":"","firstName":"Debasis","middleName":"","lastName":"Das","suffix":""},{"id":639210384,"identity":"4108676c-bdb3-43c3-8c8f-92e80cdc78f9","order_by":2,"name":"Gozzal Eshniyazova","email":"","orcid":"","institution":"Tashkent University of Information Technology","correspondingAuthor":false,"prefix":"","firstName":"Gozzal","middleName":"","lastName":"Eshniyazova","suffix":""},{"id":639210385,"identity":"c83a6508-5e62-4e40-ae82-5360a358a607","order_by":3,"name":"Doston Khasanov","email":"","orcid":"","institution":"Tashkent University of Information Technology","correspondingAuthor":false,"prefix":"","firstName":"Doston","middleName":"","lastName":"Khasanov","suffix":""},{"id":639210386,"identity":"17463f8b-b62e-40cf-ad0e-32e8dc393063","order_by":4,"name":"Halim Khujamatov","email":"","orcid":"","institution":"Tashkent University of Information Technology","correspondingAuthor":false,"prefix":"","firstName":"Halim","middleName":"","lastName":"Khujamatov","suffix":""}],"badges":[],"createdAt":"2026-04-28 04:38:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9548238/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9548238/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109206043,"identity":"a51c1acb-89c1-4abf-9ad0-6bea584fb263","added_by":"auto","created_at":"2026-05-13 15:10:46","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":354293,"visible":true,"origin":"","legend":"","description":"","filename":"RafaleResourceAwareFederatedKnowledgeDistillationonHeterogeneousClientsUpdatedNew2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9548238/v1_covered_83f7037a-b496-437e-b24b-0c7d307bd4ed.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Rafale: Towards the Efficient Resource-Aware Federated Knowledge Distillation on Heterogeneous Clients","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":"Federated Learning, Knowledge Distillation, Resource-Aware Client Clustering, Model Aggregation, Communication Efficiency","lastPublishedDoi":"10.21203/rs.3.rs-9548238/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9548238/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Knowledge distillation is an highly effective machine learning technique for transferring knowledge from a large model, i.e., teacher, to a smaller model, i.e., student, improving model efficiency without sacrificing performance. This paper proposes a novel resource-aware federated learning framework \\textit{Rafale} that incorporates an adaptive client clustering mechanism based on computational resource availability and model complexity. High-resource clients train complex teacher models, and low-resource clients leverage distilled knowledge from teachers to train lightweight student models. Further, to optimize the aggregation process by introducing a clustering-based weight-sharing mechanism that minimizes communication overhead while maintaining model accuracy. In this paper, a revolutionary Rafale framework that greatly reduces communication cost and separates model training from architectural constraints is proposed. To increase scalability and robustness, the suggested method makes use of resource-aware client selection, adaptive information sharing, and effective distillation techniques. In comparison to state-of-the-art FL approaches, experimental data show that RA-FKD delivers greater accuracy, decreases communication costs by up to 76\\%, and improves system efficiency. A cosine similarity loss \\textit{KDC} is introduced in knowledge distillation to improve the alignment between teacher and student models, enhancing knowledge transfer. Experimental evaluations on CIFAR-10 and CIFAR-100 datasets demonstrate that proposed framework achieving student test accuracy of 91.15\\% with reduction of 76.08\\% communication cost.","manuscriptTitle":"Rafale: Towards the Efficient Resource-Aware Federated Knowledge Distillation on Heterogeneous Clients","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-13 06:30:08","doi":"10.21203/rs.3.rs-9548238/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":"8cc5fa47-79e3-427b-be18-d7ceec39aa92","owner":[],"postedDate":"May 13th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-05-13T06:30:08+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-13 06:30:08","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9548238","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9548238","identity":"rs-9548238","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}