{"id":"https://openalex.org/W7162099293","doi":"https://doi.org/10.48550/arxiv.2605.21938","title":"Optimal Guarantees for Auditing R\u00e9nyi Differentially Private Machine Learning","display_name":"Optimal Guarantees for Auditing R\u00e9nyi Differentially Private Machine Learning","publication_year":2026,"publication_date":"2026-05-21","ids":{"openalex":"https://openalex.org/W7162099293","doi":"https://doi.org/10.48550/arxiv.2605.21938"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.21938","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.21938","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.21938","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136729385","display_name":"Benjamin D. Kim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Benjamin D.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136796540","display_name":"Lav R. Varshney","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Varshney, Lav R.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136745998","display_name":"Daniel Alabi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Alabi, Daniel","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":0.9545000195503235,"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":0.9545000195503235,"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":0.018300000578165054,"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":0.006800000090152025,"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/audit","display_name":"Audit","score":0.8044999837875366},{"id":"https://openalex.org/keywords/minimax","display_name":"Minimax","score":0.7757999897003174},{"id":"https://openalex.org/keywords/differential-privacy","display_name":"Differential privacy","score":0.6592000126838684},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.6115999817848206},{"id":"https://openalex.org/keywords/range","display_name":"Range (aeronautics)","score":0.49559998512268066},{"id":"https://openalex.org/keywords/divergence","display_name":"Divergence (linguistics)","score":0.40400001406669617},{"id":"https://openalex.org/keywords/logarithm","display_name":"Logarithm","score":0.38830000162124634},{"id":"https://openalex.org/keywords/mnist-database","display_name":"MNIST database","score":0.3598000109195709}],"concepts":[{"id":"https://openalex.org/C199521495","wikidata":"https://www.wikidata.org/wiki/Q181487","display_name":"Audit","level":2,"score":0.8044999837875366},{"id":"https://openalex.org/C149728462","wikidata":"https://www.wikidata.org/wiki/Q751319","display_name":"Minimax","level":2,"score":0.7757999897003174},{"id":"https://openalex.org/C23130292","wikidata":"https://www.wikidata.org/wiki/Q5275358","display_name":"Differential privacy","level":2,"score":0.6592000126838684},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6521000266075134},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.6115999817848206},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.49559998512268066},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.47189998626708984},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.453000009059906},{"id":"https://openalex.org/C207390915","wikidata":"https://www.wikidata.org/wiki/Q1230525","display_name":"Divergence (linguistics)","level":2,"score":0.40400001406669617},{"id":"https://openalex.org/C39927690","wikidata":"https://www.wikidata.org/wiki/Q11197","display_name":"Logarithm","level":2,"score":0.38830000162124634},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3702999949455261},{"id":"https://openalex.org/C190502265","wikidata":"https://www.wikidata.org/wiki/Q17069496","display_name":"MNIST database","level":3,"score":0.3598000109195709},{"id":"https://openalex.org/C99221444","wikidata":"https://www.wikidata.org/wiki/Q1532069","display_name":"Private information retrieval","level":2,"score":0.3456999957561493},{"id":"https://openalex.org/C39358052","wikidata":"https://www.wikidata.org/wiki/Q2578632","display_name":"Information security audit","level":5,"score":0.3278000056743622},{"id":"https://openalex.org/C123201435","wikidata":"https://www.wikidata.org/wiki/Q456632","display_name":"Information privacy","level":2,"score":0.3109999895095825},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.3084999918937683},{"id":"https://openalex.org/C176605952","wikidata":"https://www.wikidata.org/wiki/Q827674","display_name":"Phasor","level":4,"score":0.289000004529953},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.28209999203681946},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.28139999508857727},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.2662000060081482},{"id":"https://openalex.org/C107321475","wikidata":"https://www.wikidata.org/wiki/Q5374254","display_name":"Empirical risk minimization","level":2,"score":0.26350000500679016},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.26170000433921814},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.25450000166893005},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.25099998712539673}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.21938","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.21938","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.21938","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.21938","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":"Preprint"},"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","score":0.7357418537139893,"id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0,17,63],"study":[1],"black-box":[2,104,134],"auditing":[3,20,50,88,99,145],"for":[4,48,87,98],"machine":[5],"learning":[6],"algorithms":[7],"that":[8,26],"claim":[9],"R":[10],"\\":[11],"'enyi":[12],"differential":[13],"privacy":[14,61,116],"(RDP)":[15],"guarantees.":[16],"introduce":[18],"an":[19],"framework,":[21],"based":[22],"on":[23,125],"hypothesis":[24],"testing,":[25],"directly":[27],"estimates":[28],"R\u00e9nyi":[29,141],"divergence":[30],"between":[31],"neighboring":[32],"executions":[33],"using":[34],"the":[35,83],"Donsker-Varadhan":[36],"(DV)":[37],"variational":[38],"estimator.":[39],"Our":[40],"analysis":[41],"yields":[42],"explicit":[43],"and":[44,108,110,139],"non-asymptotic":[45],"confidence":[46],"intervals":[47],"RDP":[49,89,127],"via":[51,90],"class-restricted":[52],"DV":[53,91],"estimators,":[54],"separating":[55],"statistical":[56],"estimation":[57],"error":[58],"from":[59],"algorithmic":[60],"leakage.":[62],"prove":[64],"matching":[65],"minimax":[66],"lower":[67,128],"bounds":[68,129],"showing":[69],"that,":[70],"up":[71],"to":[72,131],"logarithmic":[73],"factors,":[74],"our":[75,96,118],"sample-complexity":[76],"guarantees":[77,86],"are":[78],"information-theoretically":[79],"optimal,":[80],"thereby":[81],"establishing":[82],"first":[84],"optimal":[85],"estimators.":[92],"Empirically,":[93],"we":[94],"instantiate":[95],"framework":[97],"DP-SGD":[100],"in":[101],"a":[102,112,121],"fully":[103],"setting.":[105],"Across":[106],"MNIST":[107],"CIFAR-10,":[109],"over":[111],"wide":[113],"range":[114],"of":[115],"regimes,":[117],"auditors":[119],"produce":[120],"strong":[122],"overall":[123],"improvement":[124],"empirical":[126],"compared":[130],"prior":[132],"state-of-the-art":[133],"methods":[135],"especially":[136],"at":[137],"small":[138],"moderate":[140],"orders":[142],"where":[143],"accurate":[144],"is":[146],"most":[147],"challenging.":[148]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-23T00:00:00"}