{"id":"https://openalex.org/W4410609104","doi":"https://doi.org/10.1109/satml64287.2025.00054","title":"Choosing Public Datasets for Private Machine Learning via Gradient Subspace Distance","display_name":"Choosing Public Datasets for Private Machine Learning via Gradient Subspace Distance","publication_year":2025,"publication_date":"2025-04-09","ids":{"openalex":"https://openalex.org/W4410609104","doi":"https://doi.org/10.1109/satml64287.2025.00054"},"language":"en","primary_location":{"id":"doi:10.1109/satml64287.2025.00054","is_oa":false,"landing_page_url":"https://doi.org/10.1109/satml64287.2025.00054","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Conference on Secure and Trustworthy Machine Learning (SaTML)","raw_type":"proceedings-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/A5057558573","display_name":"Xin Gu","orcid":null},"institutions":[{"id":"https://openalex.org/I130769515","display_name":"Pennsylvania State University","ror":"https://ror.org/04p491231","country_code":"US","type":"education","lineage":["https://openalex.org/I130769515"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Xin Gu","raw_affiliation_strings":["Penn State University"],"affiliations":[{"raw_affiliation_string":"Penn State University","institution_ids":["https://openalex.org/I130769515"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047778197","display_name":"Gautam Kamath","orcid":"https://orcid.org/0000-0003-0048-2559"},"institutions":[{"id":"https://openalex.org/I151746483","display_name":"University of Waterloo","ror":"https://ror.org/01aff2v68","country_code":"CA","type":"education","lineage":["https://openalex.org/I151746483"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Gautam Kamath","raw_affiliation_strings":["University of Waterloo"],"affiliations":[{"raw_affiliation_string":"University of Waterloo","institution_ids":["https://openalex.org/I151746483"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5001070941","display_name":"Zhiwei Steven Wu","orcid":"https://orcid.org/0000-0002-8125-8227"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhiwei Steven Wu","raw_affiliation_strings":["Carnegie Mellon University"],"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5057558573"],"corresponding_institution_ids":["https://openalex.org/I130769515"],"apc_list":null,"apc_paid":null,"fwci":2.8843,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.90988495,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"879","last_page":"900"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9998999834060669,"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":0.9998999834060669,"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/T10237","display_name":"Cryptography and Data Security","score":0.9954000115394592,"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/T10270","display_name":"Blockchain Technology Applications and Security","score":0.9574999809265137,"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/subspace-topology","display_name":"Subspace topology","score":0.8136217594146729},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.645810604095459},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5263123512268066},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3948647975921631}],"concepts":[{"id":"https://openalex.org/C32834561","wikidata":"https://www.wikidata.org/wiki/Q660730","display_name":"Subspace topology","level":2,"score":0.8136217594146729},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.645810604095459},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5263123512268066},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3948647975921631}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/satml64287.2025.00054","is_oa":false,"landing_page_url":"https://doi.org/10.1109/satml64287.2025.00054","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Conference on Secure and Trustworthy Machine Learning (SaTML)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/17","score":0.41999998688697815,"display_name":"Partnerships for the goals"}],"awards":[{"id":"https://openalex.org/G1448736033","display_name":null,"funder_award_id":"2339775,2232693","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":63,"referenced_works":["https://openalex.org/W8105021","https://openalex.org/W1731081199","https://openalex.org/W1873763122","https://openalex.org/W1970377488","https://openalex.org/W1985511977","https://openalex.org/W1992926795","https://openalex.org/W1997690112","https://openalex.org/W2004026774","https://openalex.org/W2035469609","https://openalex.org/W2045512849","https://openalex.org/W2051267297","https://openalex.org/W2104094955","https://openalex.org/W2108598243","https://openalex.org/W2131953535","https://openalex.org/W2140496876","https://openalex.org/W2194775991","https://openalex.org/W2473418344","https://openalex.org/W2488793974","https://openalex.org/W2533598788","https://openalex.org/W2535690855","https://openalex.org/W2611650229","https://openalex.org/W2750384547","https://openalex.org/W2786808285","https://openalex.org/W2788633781","https://openalex.org/W2794825826","https://openalex.org/W2797977484","https://openalex.org/W2809607392","https://openalex.org/W2930926105","https://openalex.org/W2963446712","https://openalex.org/W2963466845","https://openalex.org/W2990761674","https://openalex.org/W3031885728","https://openalex.org/W3034238904","https://openalex.org/W3175645569","https://openalex.org/W3213051863","https://openalex.org/W4309935983","https://openalex.org/W4311555033","https://openalex.org/W4312258136","https://openalex.org/W4312266295","https://openalex.org/W4399971973","https://openalex.org/W6681585023","https://openalex.org/W6752654261","https://openalex.org/W6756630351","https://openalex.org/W6759579507","https://openalex.org/W6761507452","https://openalex.org/W6770432743","https://openalex.org/W6773362992","https://openalex.org/W6778883912","https://openalex.org/W6781452205","https://openalex.org/W6787335730","https://openalex.org/W6788125050","https://openalex.org/W6796581206","https://openalex.org/W6796710205","https://openalex.org/W6801216613","https://openalex.org/W6802187784","https://openalex.org/W6802709103","https://openalex.org/W6804173803","https://openalex.org/W6810003945","https://openalex.org/W6810308698","https://openalex.org/W6811442867","https://openalex.org/W6838402796","https://openalex.org/W6839294834","https://openalex.org/W6850576105"],"related_works":["https://openalex.org/W2961085424","https://openalex.org/W4306674287","https://openalex.org/W4387369504","https://openalex.org/W4394896187","https://openalex.org/W3170094116","https://openalex.org/W4386462264","https://openalex.org/W3107602296","https://openalex.org/W4364306694","https://openalex.org/W4312192474","https://openalex.org/W4283697347"],"abstract_inverted_index":{"Differentially":[0],"private":[1,39,73,99,117],"stochastic":[2],"gradient":[3],"descent":[4],"privatizes":[5],"model":[6,23,110,122],"training":[7],"by":[8,34,42,49],"injecting":[9],"noise":[10,16,33],"into":[11],"each":[12],"iteration,":[13],"where":[14],"the":[15,20,32,50,79,96,114,127,146,165],"magnitude":[17],"increases":[18],"with":[19,108,150],"number":[21],"of":[22,57,95],"parameters.":[24],"Recent":[25],"works":[26],"suggest":[27],"that":[28,104,145],"we":[29],"can":[30,131],"reduce":[31],"leveraging":[35],"public":[36,51,58,97,115,138],"data":[37],"for":[38,70],"machine":[40],"learning,":[41],"projecting":[43],"gradients":[44,94],"onto":[45],"a":[46,55,71,84,89],"subspace":[47,91,152],"prescribed":[48],"data.":[52],"However,":[53],"given":[54],"choice":[56],"datasets,":[59],"it":[60,105],"is":[61,106,124,156],"unclear":[62],"why":[63],"certain":[64],"datasets":[65],"perform":[66],"better":[67],"than":[68],"others":[69],"particular":[72],"task,":[74],"or":[75],"how":[76],"to":[77,134,158,162],"identify":[78],"best":[80],"one.":[81],"We":[82,101,140],"provide":[83,141],"simple":[85],"metric":[86],"which":[87],"measures":[88],"low-dimensional":[90],"distance":[92,155],"between":[93],"and":[98,116,129,160],"examples.":[100],"empirically":[102],"demonstrate":[103],"well-correlated":[107],"resulting":[109],"utility":[111],"when":[112],"using":[113],"dataset":[118],"pair":[119],"(i.e.,":[120],"trained":[121],"accuracy":[123],"monotone":[125],"in":[126,164],"distance),":[128],"thus":[130],"be":[132],"used":[133],"select":[135],"an":[136],"appropriate":[137],"dataset.":[139],"theoretical":[142],"analysis":[143],"demonstrating":[144],"excess":[147],"risk":[148],"scales":[149],"this":[151],"distance.":[153],"This":[154],"easy":[157],"compute":[159],"robust":[161],"modifications":[163],"setting.":[166]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}