{"id":"https://openalex.org/W4320067890","doi":"https://doi.org/10.1145/3545008.3545073","title":"FedClassAvg: Local Representation Learning for Personalized Federated Learning on Heterogeneous Neural Networks","display_name":"FedClassAvg: Local Representation Learning for Personalized Federated Learning on Heterogeneous Neural Networks","publication_year":2022,"publication_date":"2022-08-29","ids":{"openalex":"https://openalex.org/W4320067890","doi":"https://doi.org/10.1145/3545008.3545073"},"language":"en","primary_location":{"id":"doi:10.1145/3545008.3545073","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3545008.3545073","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 51st International Conference on Parallel Processing","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102813831","display_name":"Jaehee Jang","orcid":"https://orcid.org/0000-0003-0322-5654"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Jaehee Jang","raw_affiliation_strings":["Department of Electrical and Computer Engineering, Seoul National University, Republic of Korea"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Seoul National University, Republic of Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020406135","display_name":"Heoneok Ha","orcid":null},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Heoneok Ha","raw_affiliation_strings":["Department of Electrical and Computer Engineering, Seoul National University, Republic of Korea"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Seoul National University, Republic of Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010990289","display_name":"Dahuin Jung","orcid":"https://orcid.org/0000-0002-1344-1054"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Dahuin Jung","raw_affiliation_strings":["Department of Electrical and Computer Engineering, Seoul National University, Republic of Korea"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Seoul National University, Republic of Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5086877012","display_name":"Sungroh Yoon","orcid":"https://orcid.org/0000-0002-2367-197X"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Sungroh Yoon","raw_affiliation_strings":["Department of Electrical and Computer Engineering and Interdisciplinary Program in Artificial Intelligence, Seoul National University, Republic of Korea and AIIS, ASRI, and INMC, Seoul National University, Republic of Korea"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering and Interdisciplinary Program in Artificial Intelligence, Seoul National University, Republic of Korea and AIIS, ASRI, and INMC, Seoul National University, Republic of Korea","institution_ids":["https://openalex.org/I139264467"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5102813831"],"corresponding_institution_ids":["https://openalex.org/I139264467"],"apc_list":null,"apc_paid":null,"fwci":5.3795,"has_fulltext":false,"cited_by_count":42,"citation_normalized_percentile":{"value":0.962416,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"10"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9991000294685364,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9991000294685364,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9847000241279602,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T13702","display_name":"Machine Learning in Healthcare","score":0.9431999921798706,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8555196523666382},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.6435636281967163},{"id":"https://openalex.org/keywords/extractor","display_name":"Extractor","score":0.623075008392334},{"id":"https://openalex.org/keywords/classifier","display_name":"Classifier (UML)","score":0.6205540299415588},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.613746702671051},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5884878039360046},{"id":"https://openalex.org/keywords/feature-learning","display_name":"Feature learning","score":0.5018138885498047},{"id":"https://openalex.org/keywords/federated-learning","display_name":"Federated learning","score":0.4384244978427887},{"id":"https://openalex.org/keywords/transfer-of-learning","display_name":"Transfer of learning","score":0.41290000081062317}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8555196523666382},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6435636281967163},{"id":"https://openalex.org/C117978034","wikidata":"https://www.wikidata.org/wiki/Q5422192","display_name":"Extractor","level":2,"score":0.623075008392334},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.6205540299415588},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.613746702671051},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5884878039360046},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.5018138885498047},{"id":"https://openalex.org/C2992525071","wikidata":"https://www.wikidata.org/wiki/Q50818671","display_name":"Federated learning","level":2,"score":0.4384244978427887},{"id":"https://openalex.org/C150899416","wikidata":"https://www.wikidata.org/wiki/Q1820378","display_name":"Transfer of learning","level":2,"score":0.41290000081062317},{"id":"https://openalex.org/C21880701","wikidata":"https://www.wikidata.org/wiki/Q2144042","display_name":"Process engineering","level":1,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3545008.3545073","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3545008.3545073","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 51st International Conference on Parallel Processing","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","score":0.5299999713897705,"id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W2097117768","https://openalex.org/W2187089797","https://openalex.org/W2194775991","https://openalex.org/W2883780447","https://openalex.org/W2995022099","https://openalex.org/W3035453001","https://openalex.org/W3035524453","https://openalex.org/W3094163844","https://openalex.org/W3133814152","https://openalex.org/W3168256142","https://openalex.org/W3171007011","https://openalex.org/W4287812705","https://openalex.org/W6776700526"],"related_works":["https://openalex.org/W4298221930","https://openalex.org/W3082848404","https://openalex.org/W1979583797","https://openalex.org/W2777914285","https://openalex.org/W2945706271","https://openalex.org/W4295705264","https://openalex.org/W2114169842","https://openalex.org/W2535808783","https://openalex.org/W2440023763","https://openalex.org/W2962474440"],"abstract_inverted_index":{"Personalized":[0],"federated":[1,29,56,60,197],"learning":[2,30,57,69,112,198],"is":[3,113,160],"aimed":[4],"at":[5],"allowing":[6],"numerous":[7],"clients":[8,34,93,149],"to":[9,115,142,150],"train":[10],"personalized":[11,28,55,196],"models":[12,45],"while":[13],"participating":[14],"in":[15,18],"collaborative":[16],"training":[17],"a":[19,53,144,153],"communication-efficient":[20],"manner":[21],"without":[22],"exchanging":[23],"private":[24],"data.":[25],"However,":[26],"many":[27],"algorithms":[31,193],"assume":[32],"that":[33,92],"have":[35],"the":[36,117,122,130,134,190],"same":[37],"neural":[38,65],"network":[39],"architecture,":[40],"and":[41,75,97,120,186],"those":[42],"for":[43,67,127],"heterogeneous":[44,195],"remain":[46],"understudied.":[47],"In":[48,107],"this":[49],"study,":[50],"we":[51],"propose":[52],"novel":[54],"method":[58],"called":[59],"classifier":[61,76,80],"averaging":[62],"(FedClassAvg).":[63],"Deep":[64],"networks":[66],"supervised":[68],"tasks":[70],"consist":[71],"of":[72,136,155],"feature":[73,89,110,124],"extractor":[74],"layers.":[77],"FedClassAvg":[78,146,164,182],"aggregates":[79],"weights":[81,141],"as":[82,172],"an":[83],"agreement":[84],"on":[85,88,194],"decision":[86,118],"boundaries":[87,119],"spaces":[90],"so":[91],"with":[94,152],"not":[95,166],"independently":[96],"identically":[98],"distributed":[99],"(non-iid)":[100],"data":[101,138],"can":[102],"learn":[103],"about":[104],"scarce":[105],"labels.":[106],"addition,":[108],"local":[109,123],"representation":[111],"applied":[114],"stabilize":[116],"improve":[121],"extraction":[125],"capabilities":[126],"clients.":[128],"While":[129],"existing":[131],"methods":[132],"require":[133,167],"collection":[135],"auxiliary":[137],"or":[139],"model":[140],"generate":[143],"counterpart,":[145],"only":[147],"requires":[148,176],"communicate":[151],"couple":[154],"fully":[156],"connected":[157],"layers,":[158],"which":[159,175],"highly":[161],"communication-efficient.":[162],"Moreover,":[163],"does":[165],"extra":[168],"optimization":[169],"problems":[170],"such":[171],"knowledge":[173],"transfer,":[174],"intensive":[177],"computation":[178],"overhead.":[179],"We":[180],"evaluated":[181],"through":[183],"extensive":[184],"experiments":[185],"demonstrated":[187],"it":[188],"outperforms":[189],"current":[191],"state-of-the-art":[192],"tasks.":[199]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":21},{"year":2024,"cited_by_count":14},{"year":2023,"cited_by_count":4}],"updated_date":"2026-03-27T14:29:43.386196","created_date":"2025-10-10T00:00:00"}