{"id":"https://openalex.org/W7138306887","doi":"https://doi.org/10.48550/arxiv.2603.13795","title":"Computation and Communication Efficient Federated Unlearning via On-server Gradient Conflict Mitigation and Expression","display_name":"Computation and Communication Efficient Federated Unlearning via On-server Gradient Conflict Mitigation and Expression","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7138306887","doi":"https://doi.org/10.48550/arxiv.2603.13795"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.13795","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.13795","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.13795","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129745011","display_name":"Minh-Duong Nguyen","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Nguyen, Minh-Duong","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129675850","display_name":"Senura Hansaja","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hansaja, Senura","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036902350","display_name":"Le-Tuan Duy Nguyen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nguyen, Le-Tuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122752093","display_name":"Quoc-Viet Pham","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pham, Quoc-Viet","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129708493","display_name":"Ken-Tye Yong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yong, Ken-Tye","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129678457","display_name":"Nguyen H. Tran","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tran, Nguyen H.","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5129720540","display_name":"Dung D. Le","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Le, Dung D.","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5129745011"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.21899999678134918,"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":0.21899999678134918,"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.14300000667572021,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T14347","display_name":"Big Data and Digital Economy","score":0.05849999934434891,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/key","display_name":"Key (lock)","score":0.6721000075340271},{"id":"https://openalex.org/keywords/federated-learning","display_name":"Federated learning","score":0.6172000169754028},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5702999830245972},{"id":"https://openalex.org/keywords/retraining","display_name":"Retraining","score":0.5449000000953674},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.5263000130653381},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4668999910354614}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7954000234603882},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.6721000075340271},{"id":"https://openalex.org/C2992525071","wikidata":"https://www.wikidata.org/wiki/Q50818671","display_name":"Federated learning","level":2,"score":0.6172000169754028},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5702999830245972},{"id":"https://openalex.org/C2778712577","wikidata":"https://www.wikidata.org/wiki/Q3505966","display_name":"Retraining","level":2,"score":0.5449000000953674},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.5263000130653381},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4668999910354614},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.42730000615119934},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42309999465942383},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.3540000021457672},{"id":"https://openalex.org/C2780898871","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Performance metric","level":2,"score":0.3264000117778778},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3158000111579895},{"id":"https://openalex.org/C158156997","wikidata":"https://www.wikidata.org/wiki/Q1416645","display_name":"Models of communication","level":2,"score":0.28839999437332153},{"id":"https://openalex.org/C123201435","wikidata":"https://www.wikidata.org/wiki/Q456632","display_name":"Information privacy","level":2,"score":0.2872999906539917},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.28369998931884766},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.27649998664855957},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2639999985694885},{"id":"https://openalex.org/C90559484","wikidata":"https://www.wikidata.org/wiki/Q778379","display_name":"Expression (computer science)","level":2,"score":0.2596000134944916}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.13795","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.13795","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.13795","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.13795","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[{"score":0.6075311899185181,"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Federated":[0,13,55],"Unlearning":[1,57],"(FUL)":[2],"aims":[3,109],"to":[4,35,110,121],"remove":[5],"specific":[6],"participants'":[7],"data":[8,18,134],"contributions":[9],"from":[10],"a":[11,59,72,132,140,157],"trained":[12],"Learning":[14],"model,":[15],"thereby":[16,124],"ensuring":[17],"privacy":[19],"and":[20,43,46,81,95,128,143,206],"compliance":[21],"with":[22,87,198],"regulatory":[23],"requirements.":[24],"Despite":[25],"its":[26],"potential,":[27],"progress":[28],"in":[29,189],"FUL":[30],"has":[31],"been":[32],"limited":[33],"due":[34],"several":[36],"challenges,":[37,52],"including":[38],"the":[39,78,83,88,97,100,112,116,150,167,187],"cross-client":[40],"knowledge":[41,106],"inaccessibility":[42],"high":[44],"computational":[45],"communication":[47,205],"costs.":[48,208],"To":[49,148],"overcome":[50],"these":[51],"we":[53,155],"propose":[54,156],"On-server":[56],"(FOUL),":[58],"novel":[60,158],"framework":[61],"that":[62,184],"comprises":[63],"two":[64],"key":[65,84],"stages.":[66],"The":[67],"learning-to-unlearn":[68],"stage":[69,92],"serves":[70],"as":[71],"preparatory":[73],"learning":[74],"phase,":[75],"during":[76],"which":[77,138,163],"model":[79,168],"identifies":[80],"encodes":[82],"features":[85],"associated":[86],"forget":[89],"clients.":[90],"This":[91],"is":[93],"communication-efficient":[94],"establishes":[96],"basis":[98],"for":[99,136],"subsequent":[101],"unlearning":[102,113,171,181],"process.":[103],"Subsequently,":[104],"on-server":[105],"aggregation":[107],"phase":[108],"perform":[111],"process":[114],"at":[115],"server":[117],"without":[118],"requiring":[119],"access":[120],"client":[122],"data,":[123],"preserving":[125],"both":[126],"efficiency":[127],"privacy.":[129],"We":[130],"introduce":[131],"new":[133],"setting":[135],"FUL,":[137],"enables":[139],"more":[141],"transparent":[142],"rigorous":[144],"evaluation":[145,159],"of":[146,152],"unlearning.":[147],"highlight":[149],"effectiveness":[151],"our":[153],"approach,":[154],"metric":[160],"termed":[161],"time-to-forget,":[162,201],"measures":[164],"how":[165],"quickly":[166],"achieves":[169,193],"optimal":[170],"performance.":[172],"Extensive":[173],"experiments":[174],"conducted":[175],"on":[176],"three":[177],"datasets":[178],"under":[179],"various":[180],"scenarios":[182],"demonstrate":[183],"FOUL":[185,192],"outperforms":[186],"Retraining":[188],"FUL.":[190],"Moreover,":[191],"competitive":[194],"or":[195],"superior":[196],"results":[197],"significantly":[199],"reduced":[200],"while":[202],"maintaining":[203],"low":[204],"computation":[207]},"counts_by_year":[],"updated_date":"2026-03-18T06:31:55.123368","created_date":"2026-03-18T00:00:00"}