{"id":"https://openalex.org/W4392153198","doi":"https://doi.org/10.1109/globecom54140.2023.10437587","title":"AFL: Attention-Based Byzantine-Robust Federated Learning with Vector Filter","display_name":"AFL: Attention-Based Byzantine-Robust Federated Learning with Vector Filter","publication_year":2023,"publication_date":"2023-12-04","ids":{"openalex":"https://openalex.org/W4392153198","doi":"https://doi.org/10.1109/globecom54140.2023.10437587"},"language":"en","primary_location":{"id":"doi:10.1109/globecom54140.2023.10437587","is_oa":false,"landing_page_url":"https://doi.org/10.1109/globecom54140.2023.10437587","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"GLOBECOM 2023 - 2023 IEEE Global Communications Conference","raw_type":"proceedings-article"},"type":"conference-paper","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/A5100353560","display_name":"Hao Chen","orcid":"https://orcid.org/0000-0002-5035-0063"},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Chen","raw_affiliation_strings":["School of Cyber Engineering, Xidian University,Xi'an,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Cyber Engineering, Xidian University,Xi'an,China","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059296628","display_name":"Xixiang Lv","orcid":"https://orcid.org/0000-0003-2879-241X"},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xixiang Lv","raw_affiliation_strings":["Xidian University,National Key Lab of ISN,Xi'an,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Xidian University,National Key Lab of ISN,Xi'an,China","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100401082","display_name":"Wei Zheng","orcid":"https://orcid.org/0000-0002-0666-521X"},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Zheng","raw_affiliation_strings":["School of Cyber Engineering, Xidian University,Xi'an,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Cyber Engineering, Xidian University,Xi'an,China","institution_ids":["https://openalex.org/I149594827"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I149594827"],"apc_list":null,"apc_paid":null,"fwci":0.4311,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.63750822,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"595","last_page":"600"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9998000264167786,"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":0.9998000264167786,"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/T11598","display_name":"Internet Traffic Analysis and Secure E-voting","score":0.979200005531311,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9643999934196472,"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":0.6782139539718628},{"id":"https://openalex.org/keywords/filter","display_name":"Filter (signal processing)","score":0.5719087719917297},{"id":"https://openalex.org/keywords/byzantine-architecture","display_name":"Byzantine architecture","score":0.49820470809936523},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4503933787345886},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.33228692412376404},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.18868064880371094},{"id":"https://openalex.org/keywords/history","display_name":"History","score":0.07041436433792114}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6782139539718628},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.5719087719917297},{"id":"https://openalex.org/C104562893","wikidata":"https://www.wikidata.org/wiki/Q47591","display_name":"Byzantine architecture","level":2,"score":0.49820470809936523},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4503933787345886},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.33228692412376404},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.18868064880371094},{"id":"https://openalex.org/C95457728","wikidata":"https://www.wikidata.org/wiki/Q309","display_name":"History","level":0,"score":0.07041436433792114},{"id":"https://openalex.org/C195244886","wikidata":"https://www.wikidata.org/wiki/Q41493","display_name":"Ancient history","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/globecom54140.2023.10437587","is_oa":false,"landing_page_url":"https://doi.org/10.1109/globecom54140.2023.10437587","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"GLOBECOM 2023 - 2023 IEEE Global Communications Conference","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":20,"referenced_works":["https://openalex.org/W6908809","https://openalex.org/W398859631","https://openalex.org/W2788816110","https://openalex.org/W2789903762","https://openalex.org/W2903356604","https://openalex.org/W3115472659","https://openalex.org/W3198262235","https://openalex.org/W4252654521","https://openalex.org/W4288277757","https://openalex.org/W4288578176","https://openalex.org/W6676935882","https://openalex.org/W6743821447","https://openalex.org/W6748019269","https://openalex.org/W6748626757","https://openalex.org/W6748786018","https://openalex.org/W6752600739","https://openalex.org/W6758201434","https://openalex.org/W6766633873","https://openalex.org/W6768324076","https://openalex.org/W6770634426"],"related_works":["https://openalex.org/W2961085424","https://openalex.org/W4306674287","https://openalex.org/W3046775127","https://openalex.org/W4394896187","https://openalex.org/W3170094116","https://openalex.org/W4386462264","https://openalex.org/W3107602296","https://openalex.org/W4364306694","https://openalex.org/W4312192474","https://openalex.org/W4283697347"],"abstract_inverted_index":{"Federated":[0,68],"Learning":[1,69],"(FL)":[2],"solves":[3],"the":[4,19,30,33,131,144,151,156,160],"privacy":[5],"leakage":[6],"problem":[7],"of":[8,32,134],"cross-domain":[9],"training,":[10],"but":[11],"is":[12,75],"vulnerable":[13],"to":[14,77,81],"Byzantine":[15,90,170],"attacks,":[16,181],"that":[17,28,155],"is,":[18],"adversary":[20],"uses":[21],"poisoned":[22,25],"data":[23],"or":[24,45],"local":[26],"models":[27,87],"affect":[29],"convergence":[31,115],"global":[34,113],"model.":[35],"However,":[36],"most":[37],"existing":[38],"Byzantine-robust":[39,67],"methods":[40],"either":[41],"do":[42],"not":[43],"converge,":[44],"are":[46],"only":[47],"suitable":[48],"for":[49],"honest":[50],"majority":[51],"settings":[52],"and":[53,86,102,147,150,182,186],"cannot":[54],"cope":[55],"with":[56,70,97,121,167,177],"well-designed":[57],"malicious":[58,137,169],"gradients.":[59,138],"To":[60],"this":[61],"end,":[62],"we":[63,94,125],"propose":[64,103],"a":[65,98,104,127],"Attention-based":[66],"Vector":[71],"Filter":[72],"(AFL),":[73],"which":[74,111],"1.1\u00d7":[76],"4.2\u00d7":[78],"more":[79],"resistant":[80],"attacks":[82],"on":[83],"different":[84],"datasets":[85],"than":[88],"other":[89],"defense":[91,135],"schemes.":[92],"Specifically,":[93],"bootstrap":[95],"trust":[96],"clean":[99],"root":[100],"dataset,":[101],"new":[105],"scaled":[106],"dot-product":[107],"attention-based":[108],"aggregation":[109],"method,":[110],"accelerates":[112],"model":[114,179],"while":[116],"maximally":[117],"preserving":[118],"benign":[119],"gradients":[120],"uneven":[122],"quality.":[123],"Additionally,":[124],"add":[126],"vector":[128],"filter":[129],"as":[130],"first":[132],"line":[133],"against":[136],"We":[139],"conduct":[140],"extensive":[141],"evaluations":[142],"using":[143],"MNIST,":[145],"FMNIST":[146],"CIFAR-10":[148],"datasets,":[149],"experimental":[152],"results":[153],"show":[154],"AFL":[157,173],"scheme":[158,174],"improves":[159],"average":[161],"test":[162],"accuracy,":[163],"works":[164],"well":[165],"even":[166],"80%":[168],"adversaries.":[171],"The":[172],"can":[175],"deal":[176],"state-of-the-art":[178],"poisoning":[180],"has":[183],"high":[184],"robustness":[185],"effectiveness.":[187]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-08-08T07:41:36.138363","created_date":"2025-10-10T00:00:00"}