{"id":"https://openalex.org/W7166180820","doi":"https://doi.org/10.48550/arxiv.2606.26772","title":"Escaping Iterative Parameter-Space Noise: Differentially Private Learning with a Hypernetwork","display_name":"Escaping Iterative Parameter-Space Noise: Differentially Private Learning with a Hypernetwork","publication_year":2026,"publication_date":"2026-06-25","ids":{"openalex":"https://openalex.org/W7166180820","doi":"https://doi.org/10.48550/arxiv.2606.26772"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.26772","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.26772","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.26772","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139415618","display_name":"Naoki Nishikawa","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nishikawa, Naoki","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119989939","display_name":"Shokichi Takakura","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Takakura, Shokichi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5000559398","display_name":"S. Hasegawa","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hasegawa, Satoshi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"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":0.8883000016212463,"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":0.8883000016212463,"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":0.06620000302791595,"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":0.00559999980032444,"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/noise","display_name":"Noise (video)","score":0.664900004863739},{"id":"https://openalex.org/keywords/space","display_name":"Space (punctuation)","score":0.4291999936103821},{"id":"https://openalex.org/keywords/synthetic-data","display_name":"Synthetic data","score":0.42829999327659607},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.3937999904155731},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.3741999864578247},{"id":"https://openalex.org/keywords/noisy-data","display_name":"Noisy data","score":0.3617999851703644},{"id":"https://openalex.org/keywords/artificial-noise","display_name":"Artificial noise","score":0.3582000136375427},{"id":"https://openalex.org/keywords/iterative-learning-control","display_name":"Iterative learning control","score":0.34689998626708984}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7263000011444092},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.664900004863739},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.569100022315979},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.48100000619888306},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.4291999936103821},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.42829999327659607},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3937999904155731},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3741999864578247},{"id":"https://openalex.org/C2781170535","wikidata":"https://www.wikidata.org/wiki/Q30587856","display_name":"Noisy data","level":2,"score":0.3617999851703644},{"id":"https://openalex.org/C2780909371","wikidata":"https://www.wikidata.org/wiki/Q4801092","display_name":"Artificial noise","level":4,"score":0.3582000136375427},{"id":"https://openalex.org/C117619785","wikidata":"https://www.wikidata.org/wiki/Q6094414","display_name":"Iterative learning control","level":3,"score":0.34689998626708984},{"id":"https://openalex.org/C159694833","wikidata":"https://www.wikidata.org/wiki/Q2321565","display_name":"Iterative method","level":2,"score":0.3312999904155731},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.30320000648498535},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2994999885559082},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2937000095844269},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2888999879360199},{"id":"https://openalex.org/C29265498","wikidata":"https://www.wikidata.org/wiki/Q7047719","display_name":"Noise measurement","level":3,"score":0.28200000524520874},{"id":"https://openalex.org/C73586568","wikidata":"https://www.wikidata.org/wiki/Q2600211","display_name":"Parameter space","level":2,"score":0.28200000524520874},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2768000066280365},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.26010000705718994},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.25690001249313354},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2563999891281128}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.26772","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.26772","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.26772","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.26772","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Differentially":[0],"private":[1,71],"(DP)":[2],"training":[3],"of":[4,14,52,76,133,169],"neural":[5],"networks":[6],"is":[7,83,116],"often":[8],"hindered":[9],"by":[10,17,150],"the":[11,54,74,77,89,102,105,130],"large":[12],"amount":[13],"noise":[15,27,115],"required":[16],"gradient-based":[18],"methods":[19],"such":[20],"as":[21],"DP-SGD,":[22],"which":[23],"repeatedly":[24],"inject":[25],"high-dimensional":[26],"in":[28,48,138],"parameter":[29,49],"space":[30],"throughout":[31],"training.":[32],"In":[33],"this":[34,110],"paper,":[35],"we":[36,60,162],"propose":[37],"a":[38,62,70,86,97,121,139,144],"new":[39],"framework":[40],"for":[41],"DP":[42,98],"learning":[43],"that":[44,174],"avoids":[45],"iterative":[46],"optimization":[47],"space.":[50],"Instead":[51],"updating":[53],"target":[55,78,106],"model":[56,107],"using":[57],"privatized":[58],"gradients,":[59],"employ":[61],"hypernetwork":[63,103],"trained":[64,158,182],"on":[65],"public":[66],"datasets":[67],"to":[68,73,95,166],"map":[69],"dataset":[72,99,123],"parameters":[75,108],"model.":[79],"Specifically,":[80],"each":[81],"example":[82],"embedded":[84],"into":[85,120],"low-dimensional":[87,122],"representation,":[88,124],"embeddings":[90],"are":[91],"aggregated":[92],"and":[93,101,172,185],"perturbed":[94],"obtain":[96],"embedding,":[100],"generates":[104],"from":[109],"noisy":[111],"embedding.":[112],"Because":[113],"privacy":[114,146],"injected":[117],"only":[118],"once":[119],"our":[125,151,164],"approach":[126,152,165],"can":[127],"significantly":[128],"reduce":[129],"adverse":[131],"effect":[132],"noise.":[134],"We":[135],"theoretically":[136],"show":[137,173],"synthetic":[140],"setting":[141],"that,":[142],"under":[143],"fixed":[145],"budget,":[147],"models":[148,171,181],"produced":[149],"achieve":[153],"higher":[154],"utility":[155],"than":[156,179],"those":[157],"with":[159,183],"DP-SGD.":[160],"Moreover,":[161],"apply":[163],"LoRA":[167,180],"fine-tuning":[168],"diffusion":[170],"it":[175],"achieves":[176],"lower":[177],"FID":[178],"DP-SGD":[184],"other":[186],"public-data-guided":[187],"methods.":[188]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-27T00:00:00"}