{"id":"https://openalex.org/W7126121766","doi":"https://doi.org/10.1109/tifs.2026.3659401","title":"Safeguarding Federated Learning From Data Reconstruction Attacks via Gradient Dropout","display_name":"Safeguarding Federated Learning From Data Reconstruction Attacks via Gradient Dropout","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7126121766","doi":"https://doi.org/10.1109/tifs.2026.3659401"},"language":null,"primary_location":{"id":"doi:10.1109/tifs.2026.3659401","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tifs.2026.3659401","pdf_url":null,"source":{"id":"https://openalex.org/S61310614","display_name":"IEEE Transactions on Information Forensics and Security","issn_l":"1556-6013","issn":["1556-6013","1556-6021"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Information Forensics and Security","raw_type":"journal-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/A5072699261","display_name":"Ekanut Sotthiwat","orcid":null},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":true,"raw_author_name":"Ekanut Sotthiwat","raw_affiliation_strings":["School of Computing, National University of Singapore, Queenstown, Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computing, National University of Singapore, Queenstown, Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100458187","display_name":"Chi Zhang","orcid":"https://orcid.org/0000-0002-5735-4454"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Chi Zhang","raw_affiliation_strings":["Department of Mathematics, National University of Singapore, Queenstown, Singapore"],"raw_orcid":"https://orcid.org/0000-0002-5735-4454","affiliations":[{"raw_affiliation_string":"Department of Mathematics, National University of Singapore, Queenstown, Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010903591","display_name":"Xiaokui Xiao","orcid":"https://orcid.org/0000-0003-0914-4580"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Xiaokui Xiao","raw_affiliation_strings":["School of Computing, National University of Singapore, Queenstown, Singapore"],"raw_orcid":"https://orcid.org/0000-0003-0914-4580","affiliations":[{"raw_affiliation_string":"School of Computing, National University of Singapore, Queenstown, Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5086517854","display_name":"Liangli Zhen","orcid":"https://orcid.org/0000-0003-0481-3298"},"institutions":[{"id":"https://openalex.org/I115228651","display_name":"Agency for Science, Technology and Research","ror":"https://ror.org/036wvzt09","country_code":"SG","type":"government","lineage":["https://openalex.org/I115228651"]},{"id":"https://openalex.org/I3004594783","display_name":"Institute of High Performance Computing","ror":"https://ror.org/02n0ejh50","country_code":"SG","type":"facility","lineage":["https://openalex.org/I115228651","https://openalex.org/I3004594783","https://openalex.org/I91275662"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Liangli Zhen","raw_affiliation_strings":["Institute of High Performance Computing, Agency for Science, Technology and Research (A*STAR), Fusionopolis, Singapore"],"raw_orcid":"https://orcid.org/0000-0003-0481-3298","affiliations":[{"raw_affiliation_string":"Institute of High Performance Computing, Agency for Science, Technology and Research (A*STAR), Fusionopolis, Singapore","institution_ids":["https://openalex.org/I3004594783","https://openalex.org/I115228651"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5072699261"],"corresponding_institution_ids":["https://openalex.org/I165932596"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.19012797,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"21","issue":null,"first_page":"1874","last_page":"1888"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.5602999925613403,"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":0.5602999925613403,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.40380001068115234,"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":0.0044999998062849045,"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/safeguarding","display_name":"Safeguarding","score":0.48190000653266907},{"id":"https://openalex.org/keywords/dropout","display_name":"Dropout (neural networks)","score":0.47679999470710754},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.4553999900817871},{"id":"https://openalex.org/keywords/data-sharing","display_name":"Data sharing","score":0.42100000381469727},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.36250001192092896},{"id":"https://openalex.org/keywords/data-modeling","display_name":"Data modeling","score":0.35280001163482666},{"id":"https://openalex.org/keywords/vulnerability","display_name":"Vulnerability (computing)","score":0.3513999879360199},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.3508000075817108},{"id":"https://openalex.org/keywords/gaussian-process","display_name":"Gaussian process","score":0.3402999937534332}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8255000114440918},{"id":"https://openalex.org/C2776743756","wikidata":"https://www.wikidata.org/wiki/Q5097921","display_name":"Safeguarding","level":2,"score":0.48190000653266907},{"id":"https://openalex.org/C2776145597","wikidata":"https://www.wikidata.org/wiki/Q25339462","display_name":"Dropout (neural networks)","level":2,"score":0.47679999470710754},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.4553999900817871},{"id":"https://openalex.org/C2779965156","wikidata":"https://www.wikidata.org/wiki/Q5227350","display_name":"Data sharing","level":3,"score":0.42100000381469727},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.36250001192092896},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.35280001163482666},{"id":"https://openalex.org/C95713431","wikidata":"https://www.wikidata.org/wiki/Q631425","display_name":"Vulnerability (computing)","level":2,"score":0.3513999879360199},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.3508000075817108},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.3402999937534332},{"id":"https://openalex.org/C2992525071","wikidata":"https://www.wikidata.org/wiki/Q50818671","display_name":"Federated learning","level":2,"score":0.33719998598098755},{"id":"https://openalex.org/C123201435","wikidata":"https://www.wikidata.org/wiki/Q456632","display_name":"Information privacy","level":2,"score":0.3312999904155731},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3192000091075897},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.31679999828338623},{"id":"https://openalex.org/C148730421","wikidata":"https://www.wikidata.org/wiki/Q141090","display_name":"Encryption","level":2,"score":0.30660000443458557},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.305400013923645},{"id":"https://openalex.org/C81184566","wikidata":"https://www.wikidata.org/wiki/Q1191895","display_name":"Conjugate gradient method","level":2,"score":0.305400013923645},{"id":"https://openalex.org/C72134830","wikidata":"https://www.wikidata.org/wiki/Q5166524","display_name":"Convexity","level":2,"score":0.3025999963283539},{"id":"https://openalex.org/C132964779","wikidata":"https://www.wikidata.org/wiki/Q2110223","display_name":"Raw data","level":2,"score":0.2948000133037567},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.2816999852657318},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.2782000005245209},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.27309998869895935},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.2648000121116638},{"id":"https://openalex.org/C34388435","wikidata":"https://www.wikidata.org/wiki/Q2267362","display_name":"Bounded function","level":2,"score":0.2614000141620636},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.2563000023365021},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.25589999556541443},{"id":"https://openalex.org/C2779582901","wikidata":"https://www.wikidata.org/wiki/Q21013010","display_name":"Distributed learning","level":2,"score":0.2529999911785126},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.2524999976158142},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.2522999942302704},{"id":"https://openalex.org/C23130292","wikidata":"https://www.wikidata.org/wiki/Q5275358","display_name":"Differential privacy","level":2,"score":0.2515999972820282}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tifs.2026.3659401","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tifs.2026.3659401","pdf_url":null,"source":{"id":"https://openalex.org/S61310614","display_name":"IEEE Transactions on Information Forensics and Security","issn_l":"1556-6013","issn":["1556-6013","1556-6021"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Information Forensics and Security","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G5337830919","display_name":null,"funder_award_id":"AISG4-GC-2023-007-1B","funder_id":"https://openalex.org/F4320320709","funder_display_name":"National Research Foundation Singapore"}],"funders":[{"id":"https://openalex.org/F4320320709","display_name":"National Research Foundation Singapore","ror":"https://ror.org/03cpyc314"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":44,"referenced_works":["https://openalex.org/W1473730842","https://openalex.org/W2088492763","https://openalex.org/W2108598243","https://openalex.org/W2133665775","https://openalex.org/W2473418344","https://openalex.org/W2535690855","https://openalex.org/W2591882872","https://openalex.org/W2594311007","https://openalex.org/W2611650229","https://openalex.org/W2612252966","https://openalex.org/W2783522756","https://openalex.org/W2896479683","https://openalex.org/W2930926105","https://openalex.org/W2949734902","https://openalex.org/W2962785568","https://openalex.org/W2963359731","https://openalex.org/W2963456518","https://openalex.org/W3012501605","https://openalex.org/W3016632787","https://openalex.org/W3045674654","https://openalex.org/W3084432478","https://openalex.org/W3094542121","https://openalex.org/W3175192640","https://openalex.org/W3178386862","https://openalex.org/W3189899733","https://openalex.org/W4207073300","https://openalex.org/W4213044365","https://openalex.org/W4280575929","https://openalex.org/W4285605911","https://openalex.org/W4296913515","https://openalex.org/W4310141801","https://openalex.org/W4382203360","https://openalex.org/W4382318921","https://openalex.org/W4387011206","https://openalex.org/W4390872039","https://openalex.org/W4394699048","https://openalex.org/W4399056776","https://openalex.org/W4400772252","https://openalex.org/W4402123838","https://openalex.org/W4405270168","https://openalex.org/W4406689838","https://openalex.org/W4409362364","https://openalex.org/W4412446159","https://openalex.org/W7116667914"],"related_works":[],"abstract_inverted_index":{"Federated":[0],"Learning":[1],"(FL)":[2],"enables":[3],"collaborative":[4],"model":[5,61,145,181,221],"training":[6,38,153],"across":[7,98],"distributed":[8],"participants":[9],"without":[10],"sharing":[11],"raw":[12],"data,":[13],"offering":[14],"a":[15,51,71,84,212],"privacy-preserving":[16],"paradigm.":[17],"However,":[18],"recent":[19],"studies":[20],"on":[21],"gradient":[22,86,148],"inversion":[23],"attacks":[24],"have":[25],"demonstrated":[26],"the":[27,77,102,116,133,151,194],"vulnerability":[28],"of":[29,73,101],"FL":[30],"to":[31,193],"adversaries":[32],"who":[33],"can":[34,119],"reconstruct":[35],"sensitive":[36],"local":[37],"data":[39],"from":[40,124],"shared":[41],"gradients.":[42,111],"To":[43],"mitigate":[44],"this":[45,94,140],"threat,":[46],"we":[47,137,170],"propose":[48],"Gradient":[49,64,198],"Dropout,":[50],"novel":[52],"defense":[53],"mechanism":[54,95,142],"that":[55,88,105,115,139,172],"disrupts":[56],"reconstruction":[57,91],"attempts":[58],"while":[59,180],"preserving":[60],"utility.":[62],"Specifically,":[63],"Dropout":[65,199],"perturbs":[66],"gradients":[67,118],"by":[68],"randomly":[69],"scaling":[70],"subset":[72],"components":[74],"and":[75,150,220],"replacing":[76],"remainder":[78],"with":[79,186],"Gaussian":[80],"noise,":[81],"thereby":[82,128],"creating":[83],"transformed":[85],"space":[87],"significantly":[89],"impedes":[90],"attempts.":[92],"Moreover,":[93],"is":[96,183,200],"applied":[97],"all":[99],"layers":[100],"model,":[103],"ensuring":[104],"attackers":[106],"cannot":[107],"exploit":[108],"any":[109],"unperturbed":[110],"Theoretical":[112],"analysis":[113],"reveals":[114],"perturbed":[117],"be":[120],"kept":[121],"sufficiently":[122],"distant":[123],"their":[125],"true":[126],"values,":[127],"providing":[129,211],"safety":[130],"guarantees":[131],"for":[132,206],"proposed":[134],"algorithm.":[135],"Furthermore,":[136],"demonstrate":[138],"protection":[141],"minimally":[143],"impacts":[144],"performance,":[146],"as":[147,202],"dropout":[149],"original":[152],"dynamics":[154],"remain":[155],"effectively":[156],"bounded":[157],"under":[158],"certain":[159],"convexity":[160],"conditions.":[161],"These":[162],"findings":[163],"are":[164],"substantiated":[165],"through":[166],"experimental":[167],"evaluations,":[168],"where":[169],"show":[171],"various":[173],"attack":[174],"methods":[175],"yield":[176],"low-quality":[177],"reconstructed":[178],"images":[179],"performance":[182],"largely":[184],"preserved,":[185],"less":[187],"than":[188],"2%":[189],"accuracy":[190],"reduction":[191],"relative":[192],"baseline.":[195],"As":[196],"such,":[197],"presented":[201],"an":[203],"effective":[204],"solution":[205],"safeguarding":[207],"privacy":[208,216],"in":[209],"FL,":[210],"balanced":[213],"trade-off":[214],"between":[215],"protection,":[217],"computational":[218],"efficiency,":[219],"accuracy.":[222]},"counts_by_year":[],"updated_date":"2026-02-09T05:59:30.833894","created_date":"2026-01-30T00:00:00"}