{"id":"https://openalex.org/W4387857707","doi":"https://doi.org/10.56553/popets-2024-0029","title":"SEDMA: Self-Distillation with Model Aggregation for Membership Privacy","display_name":"SEDMA: Self-Distillation with Model Aggregation for Membership Privacy","publication_year":2023,"publication_date":"2023-10-22","ids":{"openalex":"https://openalex.org/W4387857707","doi":"https://doi.org/10.56553/popets-2024-0029"},"language":"en","primary_location":{"id":"doi:10.56553/popets-2024-0029","is_oa":true,"landing_page_url":"https://doi.org/10.56553/popets-2024-0029","pdf_url":"https://petsymposium.org/popets/2024/popets-2024-0029.pdf","source":{"id":"https://openalex.org/S4210183172","display_name":"Proceedings on Privacy Enhancing Technologies","issn_l":"2299-0984","issn":["2299-0984"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320322","host_organization_name":"De Gruyter Open","host_organization_lineage":["https://openalex.org/P4310320322","https://openalex.org/P4310313990"],"host_organization_lineage_names":["De Gruyter Open","De Gruyter"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings on Privacy Enhancing Technologies","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://petsymposium.org/popets/2024/popets-2024-0029.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101126904","display_name":"Tsunato Nakai","orcid":null},"institutions":[{"id":"https://openalex.org/I4210133125","display_name":"Mitsubishi Electric (Japan)","ror":"https://ror.org/033y26782","country_code":"JP","type":"company","lineage":["https://openalex.org/I1306287861","https://openalex.org/I4210133125"]},{"id":"https://openalex.org/I4210159266","display_name":"Mitsubishi Electric (United States)","ror":"https://ror.org/053jnhe44","country_code":"US","type":"company","lineage":["https://openalex.org/I1306287861","https://openalex.org/I4210133125","https://openalex.org/I4210159266"]}],"countries":["JP","US"],"is_corresponding":false,"raw_author_name":"Tsunato Nakai","raw_affiliation_strings":["Mitsubishi Electric Corporation","Mitsubishi Electric Corporation Kamakura, Kanagawa, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Mitsubishi Electric Corporation","institution_ids":["https://openalex.org/I4210159266"]},{"raw_affiliation_string":"Mitsubishi Electric Corporation Kamakura, Kanagawa, Japan","institution_ids":["https://openalex.org/I4210133125"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100423404","display_name":"Ye Wang","orcid":"https://orcid.org/0000-0001-5220-1830"},"institutions":[{"id":"https://openalex.org/I4210159266","display_name":"Mitsubishi Electric (United States)","ror":"https://ror.org/053jnhe44","country_code":"US","type":"company","lineage":["https://openalex.org/I1306287861","https://openalex.org/I4210133125","https://openalex.org/I4210159266"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ye Wang","raw_affiliation_strings":["Mitsubishi Electric Research Laboratories","Mitsubishi Electric Research Laboratories Cambridge, MA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Mitsubishi Electric Research Laboratories","institution_ids":["https://openalex.org/I4210159266"]},{"raw_affiliation_string":"Mitsubishi Electric Research Laboratories Cambridge, MA, USA","institution_ids":["https://openalex.org/I4210159266"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063582139","display_name":"Kota Yoshida","orcid":"https://orcid.org/0000-0003-1293-6415"},"institutions":[{"id":"https://openalex.org/I135768898","display_name":"Ritsumeikan University","ror":"https://ror.org/0197nmd03","country_code":"JP","type":"education","lineage":["https://openalex.org/I135768898","https://openalex.org/I4390039241"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Kota Yoshida","raw_affiliation_strings":["Ritsumeikan University","Ritsumeikan University Kusatsu, Shiga, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ritsumeikan University","institution_ids":["https://openalex.org/I135768898"]},{"raw_affiliation_string":"Ritsumeikan University Kusatsu, Shiga, Japan","institution_ids":["https://openalex.org/I135768898"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5007179822","display_name":"Takeshi Fujino","orcid":"https://orcid.org/0000-0001-9441-3137"},"institutions":[{"id":"https://openalex.org/I135768898","display_name":"Ritsumeikan University","ror":"https://ror.org/0197nmd03","country_code":"JP","type":"education","lineage":["https://openalex.org/I135768898","https://openalex.org/I4390039241"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Takeshi Fujino","raw_affiliation_strings":["Ritsumeikan University","Ritsumeikan University Kusatsu, Shiga, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ritsumeikan University","institution_ids":["https://openalex.org/I135768898"]},{"raw_affiliation_string":"Ritsumeikan University Kusatsu, Shiga, Japan","institution_ids":["https://openalex.org/I135768898"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.6465,"has_fulltext":true,"cited_by_count":5,"citation_normalized_percentile":{"value":0.75204248,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":98},"biblio":{"volume":"2024","issue":"1","first_page":"494","last_page":"508"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9815999865531921,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9815999865531921,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9725000262260437,"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":0.7638240456581116},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6010853052139282},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5800317525863647},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.5602495670318604},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.5403615832328796},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.507089376449585},{"id":"https://openalex.org/keywords/federated-learning","display_name":"Federated learning","score":0.4713451564311981},{"id":"https://openalex.org/keywords/aggregate","display_name":"Aggregate (composite)","score":0.46691372990608215},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4615400433540344},{"id":"https://openalex.org/keywords/privacy-protection","display_name":"Privacy protection","score":0.44947993755340576},{"id":"https://openalex.org/keywords/data-mining","display_name":"Data mining","score":0.40361666679382324},{"id":"https://openalex.org/keywords/computer-security","display_name":"Computer security","score":0.15398162603378296}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7638240456581116},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6010853052139282},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5800317525863647},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5602495670318604},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.5403615832328796},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.507089376449585},{"id":"https://openalex.org/C2992525071","wikidata":"https://www.wikidata.org/wiki/Q50818671","display_name":"Federated learning","level":2,"score":0.4713451564311981},{"id":"https://openalex.org/C4679612","wikidata":"https://www.wikidata.org/wiki/Q866298","display_name":"Aggregate (composite)","level":2,"score":0.46691372990608215},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4615400433540344},{"id":"https://openalex.org/C3017597292","wikidata":"https://www.wikidata.org/wiki/Q25052250","display_name":"Privacy protection","level":2,"score":0.44947993755340576},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.40361666679382324},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.15398162603378296},{"id":"https://openalex.org/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.0},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","level":1,"score":0.0},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.56553/popets-2024-0029","is_oa":true,"landing_page_url":"https://doi.org/10.56553/popets-2024-0029","pdf_url":"https://petsymposium.org/popets/2024/popets-2024-0029.pdf","source":{"id":"https://openalex.org/S4210183172","display_name":"Proceedings on Privacy Enhancing Technologies","issn_l":"2299-0984","issn":["2299-0984"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320322","host_organization_name":"De Gruyter Open","host_organization_lineage":["https://openalex.org/P4310320322","https://openalex.org/P4310313990"],"host_organization_lineage_names":["De Gruyter Open","De Gruyter"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings on Privacy Enhancing Technologies","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.56553/popets-2024-0029","is_oa":true,"landing_page_url":"https://doi.org/10.56553/popets-2024-0029","pdf_url":"https://petsymposium.org/popets/2024/popets-2024-0029.pdf","source":{"id":"https://openalex.org/S4210183172","display_name":"Proceedings on Privacy Enhancing Technologies","issn_l":"2299-0984","issn":["2299-0984"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320322","host_organization_name":"De Gruyter Open","host_organization_lineage":["https://openalex.org/P4310320322","https://openalex.org/P4310313990"],"host_organization_lineage_names":["De Gruyter Open","De Gruyter"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings on Privacy Enhancing Technologies","raw_type":"journal-article"},"sustainable_development_goals":[{"display_name":"Partnerships for the goals","id":"https://metadata.un.org/sdg/17","score":0.4300000071525574}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4387857707.pdf"},"referenced_works_count":42,"referenced_works":["https://openalex.org/W1821462560","https://openalex.org/W2095705004","https://openalex.org/W2194775991","https://openalex.org/W2473418344","https://openalex.org/W2535690855","https://openalex.org/W2785361959","https://openalex.org/W2786233556","https://openalex.org/W2795435272","https://openalex.org/W2798657499","https://openalex.org/W2884280357","https://openalex.org/W2884943453","https://openalex.org/W2887995258","https://openalex.org/W2930926105","https://openalex.org/W2963070423","https://openalex.org/W2963378725","https://openalex.org/W2963857521","https://openalex.org/W2981338466","https://openalex.org/W3013068160","https://openalex.org/W3015625436","https://openalex.org/W3023716276","https://openalex.org/W3045469364","https://openalex.org/W3045700442","https://openalex.org/W3104224589","https://openalex.org/W3110164654","https://openalex.org/W3115042282","https://openalex.org/W3118608800","https://openalex.org/W3138758728","https://openalex.org/W3138815606","https://openalex.org/W3154109599","https://openalex.org/W3170237968","https://openalex.org/W3170901302","https://openalex.org/W3205533264","https://openalex.org/W3208549305","https://openalex.org/W3214437258","https://openalex.org/W4287553002","https://openalex.org/W4288057780","https://openalex.org/W4288103499","https://openalex.org/W4288563649","https://openalex.org/W4297799122","https://openalex.org/W4308410483","https://openalex.org/W4315706478","https://openalex.org/W4318619660"],"related_works":["https://openalex.org/W4298221930","https://openalex.org/W2777914285","https://openalex.org/W4287823391","https://openalex.org/W3013363440","https://openalex.org/W4312762663","https://openalex.org/W4317941881","https://openalex.org/W2055243143","https://openalex.org/W3035927627","https://openalex.org/W3128909129","https://openalex.org/W4280588203"],"abstract_inverted_index":{"Membership":[0],"inference":[1],"attacks":[2],"(MIAs)":[3],"are":[4],"important":[5],"measures":[6],"to":[7,72,93,128],"evaluate":[8],"potential":[9],"risks":[10],"of":[11,44,90,115,168],"privacy":[12,170],"leakage":[13],"from":[14],"machine":[15],"learning":[16],"(ML)":[17],"models.":[18],"State-of-the-art":[19],"MIA":[20,62,164,194,199],"defenses":[21,36,165],"have":[22],"achieved":[23],"favorable":[24,224],"privacy-utility":[25,216,225],"trade-offs":[26,217,226],"using":[27,69],"knowledge":[28],"distillation":[29],"on":[30,51,67,107,150],"split":[31,53,94,109],"training":[32,96,130],"datasets.":[33,54],"However,":[34],"such":[35,142],"increase":[37],"computational":[38,176,202,229],"costs":[39],"as":[40,82,143],"a":[41,60,211],"large":[42],"number":[43],"the":[45,52,74,78,95,129,133,139,192,214],"ML":[46,104,135],"models":[47,105,136],"must":[48],"be":[49],"trained":[50,106],"In":[55],"this":[56],"study,":[57],"we":[58,158],"proposed":[59],"new":[61],"defense,":[63],"called":[64],"SEDMA,":[65],"based":[66],"self-distillation":[68],"model":[70,79,119,122,140,173,187],"aggregation":[71,120],"mitigate":[73],"MIAs,":[75],"inspired":[76],"by":[77,124],"parameter":[80],"averaging":[81],"used":[83],"in":[84,144,166,196,218],"federated":[85,145],"learning.":[86,146],"The":[87,112],"key":[88],"idea":[89],"SEDMA":[91,116,161,179,204,221],"is":[92,117],"dataset":[97],"into":[98],"several":[99],"parts":[100],"and":[101,137,156,175,227],"aggregate":[102],"multiple":[103,134],"each":[108],"for":[110],"self-distillation.":[111],"intuitive":[113],"explanation":[114],"that":[118,160],"prevents":[121],"over-fitting":[123],"smoothing":[125],"information":[126],"related":[127],"data":[131],"among":[132],"preserving":[138],"utility,":[141],"Through":[147],"our":[148],"experiments":[149],"major":[151],"benchmark":[152],"datasets":[153],"(Purchase100,":[154],"Texas100,":[155],"CIFAR100),":[157],"show":[159],"outperforms":[162],"state-of-the-art":[163,197,215],"terms":[167],"membership":[169],"(MIA":[171],"accuracy),":[172],"accuracy,":[174],"costs.":[177,230],"Specifically,":[178],"incurs":[180],"at":[181],"most":[182],"approximately":[183],"3":[184],"-":[185],"5%":[186],"accuracy":[188,195],"drop,":[189],"while":[190],"achieving":[191],"lowest":[193],"empirical":[198],"defenses.":[200,220],"For":[201],"costs,":[203],"takes":[205],"significantly":[206],"less":[207],"processing":[208],"time":[209],"than":[210],"defense":[212],"with":[213],"previous":[219],"achieves":[222],"both":[223],"low":[228]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1}],"updated_date":"2026-08-11T07:18:39.950985","created_date":"2025-10-10T00:00:00"}