{"id":846901,"date":"2022-05-23T11:44:02","date_gmt":"2022-05-23T18:44:02","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/"},"modified":"2022-09-21T09:59:13","modified_gmt":"2022-09-21T16:59:13","slug":"personalized-federated-learning-with-adaptive-batchnorm-for-healthcare","status":"publish","type":"msr-research-item","link":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/personalized-federated-learning-with-adaptive-batchnorm-for-healthcare\/","title":{"rendered":"Personalized Federated Learning with Adaptive Batchnorm for Healthcare"},"content":{"rendered":"<p><span dir=\"ltr\" role=\"presentation\">There is a growing interest in applying machine learning techniques to healthcare. Recently, federated learning (FL) is <\/span><span dir=\"ltr\" role=\"presentation\">gaining popularity since it allows researchers to train powerful models without compromising data privacy and security. However, the <\/span><span dir=\"ltr\" role=\"presentation\">performance of existing FL approaches often deteriorates when encountering non-iid situations where there exist distribution gaps <\/span><span dir=\"ltr\" role=\"presentation\">among clients, and few previous efforts focus on personalization in healthcare. In this article, we propose FedAP to tackle domain shifts <\/span><span dir=\"ltr\" role=\"presentation\">and then obtain personalized models for local clients. FedAP learns the similarity between clients based on the statistics of the batch <\/span><span dir=\"ltr\" role=\"presentation\">normalization layers while preserving the specificity of each client with different local batch normalization. Comprehensive experiments <\/span><span dir=\"ltr\" role=\"presentation\">on five healthcare benchmarks demonstrate that FedAP achieves better accuracy compared to state-of-the-art methods (e.g.,<\/span> <span dir=\"ltr\" role=\"presentation\">10<\/span><span dir=\"ltr\" role=\"presentation\">%+ <\/span><span dir=\"ltr\" role=\"presentation\">accuracy improvement for PAMAP2) with faster convergence speed.<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"<p>There is a growing interest in applying machine learning techniques to healthcare. Recently, federated learning (FL) is gaining popularity since it allows researchers to train powerful models without compromising data privacy and security. However, the performance of existing FL approaches often deteriorates when encountering non-iid situations where there exist distribution gaps among clients, and few [&hellip;]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13556],"msr-publication-type":[193715],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-field-of-study":[246793,253024,247039,248329],"msr-conference":[],"msr-journal":[259135],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-846901","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-artificial-intelligence","msr-locale-en_us","msr-field-of-study-distributed-computing","msr-field-of-study-federated-learning","msr-field-of-study-health-care","msr-field-of-study-personalization"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2022-5-23","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"url","viewUrl":"false","id":"false","title":"https:\/\/arxiv.org\/pdf\/2112.00734v3.pdf","label_id":"252679","label":0},{"type":"url","viewUrl":"false","id":"false","title":"https:\/\/ieeexplore.ieee.org\/document\/9780172","label_id":"243109","label":0}],"msr_related_uploader":"","msr_attachments":[],"msr-author-ordering":[{"type":"text","value":"Wang Lu","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Jindong Wang","user_id":38455,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Jindong Wang"},{"type":"text","value":"Yiqiang Chen","user_id":0,"rest_url":false},{"type":"text","value":"Xin Qin","user_id":0,"rest_url":false},{"type":"text","value":"Renjun Xu","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Dimitrios Dimitriadis","user_id":37521,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Dimitrios Dimitriadis"},{"type":"user_nicename","value":"Tao Qin","user_id":33871,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Tao Qin"}],"msr_impact_theme":[],"msr_research_lab":[199560,199565],"msr_event":[],"msr_group":[705946,756487,761911,793670],"msr_project":[658488],"publication":[],"video":[],"download":[],"msr_publication_type":"article","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/846901"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":2,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/846901\/revisions"}],"predecessor-version":[{"id":847408,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/846901\/revisions\/847408"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=846901"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=846901"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=846901"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=846901"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=846901"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=846901"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=846901"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=846901"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=846901"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=846901"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=846901"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=846901"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=846901"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=846901"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=846901"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}