FedCEA: Efficient Adaptive Personalized Federated Learning based on Critical Learning Periods

FedCEA: Efficient Adaptive Personalized Federated Learning based on Critical Learning Periods | 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 FedCEA: Efficient Adaptive Personalized Federated Learning based on Critical Learning Periods Yichun Yu, Xiaoyi Yang, Zheping Chen, Yuqing Lan, Zhihuan Xing, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4630899/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 3 You are reading this latest preprint version Abstract Federated learning (FL) faces significant challenges due to statistical heterogeneity, which undermines the global model's generalization capability across diverse clients. Personalized FL (pFL) has been extensively studied to address this, but most existing approaches erroneously assume all stages of the FL training process are equally critical and require the participation of all clients, leading to substantial computational and communication overhead. Addressing this flaw, we propose an efficient adaptive pFL method, FedCEA, based on critical learning periods (CLP). FedCEA tailors efficient models for each client while ensuring data privacy and security by considering CLP during the training process to guide client selection. This approach reduces client-server communications, accelerating model convergence. To enhance the global model's generalization across clients with statistical heterogeneity, we introduce an Adaptive Initialization of Local Models (AILM) module with a personalized aggregation strategy. Additionally, training parameters are dynamically adjusted based on dataset quality to ensure efficiency. FedCEA also employs a compression method that assesses the importance of model parameters, reducing communication and computational costs. To evaluate the effectiveness of FedCEA, we conduct extensive experiments with four benchmark datasets in computer vision and natural language processing domains. The results consistently showed that FedCEA achieves improved accuracy and superior communication efficiency compared to state-of-the-art methods. This makes FedCEA a promising solution for handling heterogeneity in federated learning training. Code is available at https://github.com/buaaYYC/FedCEA/tree/main . Critical Learning Periods Client Selection Adaptive Initialization of Local Models Personalized Federated Learning Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editor assigned by journal 25 Jun, 2024 Submission checks completed at journal 25 Jun, 2024 First submitted to journal 24 Jun, 2024 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-4630899","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":318821947,"identity":"8737b2dc-848b-4937-ba5f-21688e168790","order_by":0,"name":"Yichun Yu","email":"","orcid":"","institution":"Beihang University","correspondingAuthor":false,"prefix":"","firstName":"Yichun","middleName":"","lastName":"Yu","suffix":""},{"id":318821948,"identity":"46ca2594-8c34-4b5e-8052-5b3971b1e172","order_by":1,"name":"Xiaoyi Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8klEQVRIiWNgGAWjYPACCR429h4o+wBxWmzk+HjOkKYlzVhOIodILQY3cgw/F/w6nNgm+faYxI8aBjm+GwmMnwvwaJGckWMsPbMPqEU6L02y5xiDseSNBGbpGXi08EvkGEjz9oC05JgBA4EhccONBDZmHjxa2CRyjH+DtUieMZP884+hnqAWoC1m0jw/0ozZJHjMpHnbGBIMCGmR7HlWZs3bYCPHxpOXbC3bJ2E488zDZml8WgyOJ2++zfNHgke+/ezBm2++2cjzHU8++BmfFgaBDAMGxjY4VwKIGRvwaQB65vgDBoY/+NWMglEwCkbBCAcAiVRIgb8Q8zoAAAAASUVORK5CYII=","orcid":"","institution":"Beihang University","correspondingAuthor":true,"prefix":"","firstName":"Xiaoyi","middleName":"","lastName":"Yang","suffix":""},{"id":318821949,"identity":"24d6659b-52bc-4057-86a7-72e25aa885bc","order_by":2,"name":"Zheping Chen","email":"","orcid":"","institution":"Beihang University","correspondingAuthor":false,"prefix":"","firstName":"Zheping","middleName":"","lastName":"Chen","suffix":""},{"id":318821950,"identity":"950d9d03-0fa0-4da0-97a5-da144fbb4add","order_by":3,"name":"Yuqing Lan","email":"","orcid":"","institution":"Beihang University","correspondingAuthor":false,"prefix":"","firstName":"Yuqing","middleName":"","lastName":"Lan","suffix":""},{"id":318821951,"identity":"f0dcd678-7bab-4817-81a0-3ce1242c0eec","order_by":4,"name":"Zhihuan Xing","email":"","orcid":"","institution":"Beihang University","correspondingAuthor":false,"prefix":"","firstName":"Zhihuan","middleName":"","lastName":"Xing","suffix":""},{"id":318821952,"identity":"942a00de-561d-479a-9108-358858bbddcb","order_by":5,"name":"Dan Yu","email":"","orcid":"","institution":"China Standard Intelligent Security","correspondingAuthor":false,"prefix":"","firstName":"Dan","middleName":"","lastName":"Yu","suffix":""}],"badges":[],"createdAt":"2024-06-24 14:33:55","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4630899/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4630899/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":60442011,"identity":"dc9cd109-128c-4617-a5ff-58c349daee1f","added_by":"auto","created_at":"2024-07-16 19:43:59","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3611621,"visible":true,"origin":"","legend":"","description":"","filename":"SpringerNatureFedCEA2406242218.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4630899/v1_covered_c07b43bf-66c5-4625-a87e-e07f7005f6ac.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"FedCEA: Efficient Adaptive Personalized Federated Learning based on Critical Learning Periods","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"the-visual-computer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"tvcj","sideBox":"Learn more about [The Visual Computer](http://link.springer.com/journal/371)","snPcode":"371","submissionUrl":"https://submission.nature.com/new-submission/371/3","title":"The Visual Computer","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Critical Learning Periods, Client Selection, Adaptive Initialization of Local Models, Personalized Federated Learning","lastPublishedDoi":"10.21203/rs.3.rs-4630899/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4630899/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Federated learning (FL) faces significant challenges due to statistical heterogeneity, which undermines the global model's generalization capability across diverse clients. Personalized FL (pFL) has been extensively studied to address this, but most existing approaches erroneously assume all stages of the FL training process are equally critical and require the participation of all clients, leading to substantial computational and communication overhead. Addressing this flaw, we propose an efficient adaptive pFL method, FedCEA, based on critical learning periods (CLP). FedCEA tailors efficient models for each client while ensuring data privacy and security by considering CLP during the training process to guide client selection. This approach reduces client-server communications, accelerating model convergence. To enhance the global model's generalization across clients with statistical heterogeneity, we introduce an Adaptive Initialization of Local Models (AILM) module with a personalized aggregation strategy. Additionally, training parameters are dynamically adjusted based on dataset quality to ensure efficiency. FedCEA also employs a compression method that assesses the importance of model parameters, reducing communication and computational costs. To evaluate the effectiveness of FedCEA, we conduct extensive experiments with four benchmark datasets in computer vision and natural language processing domains. The results consistently showed that FedCEA achieves improved accuracy and superior communication efficiency compared to state-of-the-art methods. This makes FedCEA a promising solution for handling heterogeneity in federated learning training. Code is available at https://github.com/buaaYYC/FedCEA/tree/main.","manuscriptTitle":"FedCEA: Efficient Adaptive Personalized Federated Learning based on Critical Learning Periods","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-07-16 19:35:44","doi":"10.21203/rs.3.rs-4630899/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorAssigned","content":"","date":"2024-06-25T12:07:19+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-06-25T07:26:14+00:00","index":"","fulltext":""},{"type":"submitted","content":"The Visual Computer","date":"2024-06-24T14:32:35+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"the-visual-computer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"tvcj","sideBox":"Learn more about [The Visual Computer](http://link.springer.com/journal/371)","snPcode":"371","submissionUrl":"https://submission.nature.com/new-submission/371/3","title":"The Visual Computer","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"3233d55c-cf55-4e1b-9e52-0ebec3253f3d","owner":[],"postedDate":"July 16th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2024-07-16T19:35:45+00:00","versionOfRecord":[],"versionCreatedAt":"2024-07-16 19:35:44","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4630899","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4630899","identity":"rs-4630899","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}