{"id":"https://openalex.org/W7212948709","doi":"https://doi.org/10.48550/arxiv.2609.12620","title":"Correlation-Guided Fast Machine Unlearning via Hessian Analysis","display_name":"Correlation-Guided Fast Machine Unlearning via Hessian Analysis","publication_year":2026,"publication_date":"2026-09-11","ids":{"openalex":"https://openalex.org/W7212948709","doi":"https://doi.org/10.48550/arxiv.2609.12620"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.12620","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.12620","pdf_url":"https://arxiv.org/pdf/2609.12620","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.12620","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5081361213","display_name":"Ayushi Thakur","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I56404289","display_name":"Indian Institute of Technology BHU","ror":"https://ror.org/01kh5gc44","country_code":"IN","type":"education","lineage":["https://openalex.org/I56404289","https://openalex.org/I91357014"]},{"id":"https://openalex.org/I91357014","display_name":"Banaras Hindu University","ror":"https://ror.org/04cdn2797","country_code":"IN","type":"education","lineage":["https://openalex.org/I91357014"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Thakur, Ayushi","raw_affiliation_strings":["Department of Computer Science and Engineering , Indian Institute of Technology (Banaras Hindu University) Varanasi , India"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering , Indian Institute of Technology (Banaras Hindu University) Varanasi , India","institution_ids":["https://openalex.org/I56404289","https://openalex.org/I91357014"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137022242","display_name":"Ruchir Gupta","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I56404289","display_name":"Indian Institute of Technology BHU","ror":"https://ror.org/01kh5gc44","country_code":"IN","type":"education","lineage":["https://openalex.org/I56404289","https://openalex.org/I91357014"]},{"id":"https://openalex.org/I91357014","display_name":"Banaras Hindu University","ror":"https://ror.org/04cdn2797","country_code":"IN","type":"education","lineage":["https://openalex.org/I91357014"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Gupta, Ruchir","raw_affiliation_strings":["Department of Computer Science and Engineering , Indian Institute of Technology (Banaras Hindu University) Varanasi , India"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering , Indian Institute of Technology (Banaras Hindu University) Varanasi , India","institution_ids":["https://openalex.org/I56404289","https://openalex.org/I91357014"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121818825","display_name":"Amit Kumar Jaiswal","orcid":"https://orcid.org/0000-0001-7956-8392","observed_orcids":["https://orcid.org/0000-0001-7956-8392","https://orcid.org/0000-0001-8848-7041"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jaiswal, Amit Kumar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137031240","display_name":"Prayag Tiwari","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I746986","display_name":"Halmstad University","ror":"https://ror.org/03h0qfp10","country_code":"SE","type":"education","lineage":["https://openalex.org/I746986"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Tiwari, Prayag","raw_affiliation_strings":["Halmstad University , Sweden"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Halmstad University , Sweden","institution_ids":["https://openalex.org/I746986"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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":0.9818999767303467,"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":0.9818999767303467,"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":0.016599999740719795,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.0003000000142492354,"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/machine-unlearning","display_name":"machine unlearning","score":0.9559999704360962},{"id":"https://openalex.org/keywords/hessian-analysis","display_name":"Hessian analysis","score":0.8619999885559082},{"id":"https://openalex.org/keywords/influence-function","display_name":"influence function","score":0.6549999713897705},{"id":"https://openalex.org/keywords/right-to-be-forgotten","display_name":"right to be forgotten","score":0.7789999842643738},{"id":"https://openalex.org/keywords/privacy-preserving-machine-learning","display_name":"privacy-preserving machine learning","score":0.7099999785423279},{"id":"https://openalex.org/keywords/intrusion-detection","display_name":"intrusion detection","score":0.5590000152587891},{"id":"https://openalex.org/keywords/spam-filtering","display_name":"spam filtering","score":0.7549999952316284},{"id":"https://openalex.org/keywords/threat-intelligence","display_name":"threat intelligence","score":0.7900000214576721},{"id":"https://openalex.org/keywords/membership-inference-attacks","display_name":"membership inference attacks","score":0.753000020980835},{"id":"https://openalex.org/keywords/computational-efficiency","display_name":"computational efficiency","score":0.6129999756813049}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.727400004863739},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.5335000157356262},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5097000002861023},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.4812000095844269},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.4796000123023987},{"id":"https://openalex.org/C203616005","wikidata":"https://www.wikidata.org/wiki/Q620495","display_name":"Hessian matrix","level":2,"score":0.46790000796318054},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45590001344680786},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.4172999858856201},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.4066999852657318},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.38909998536109924},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.38760000467300415},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.38749998807907104},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.3864000141620636},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3785000145435333},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.3366999924182892},{"id":"https://openalex.org/C7149132","wikidata":"https://www.wikidata.org/wiki/Q1377840","display_name":"Forgetting","level":2,"score":0.3330000042915344},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.32499998807907104},{"id":"https://openalex.org/C35525427","wikidata":"https://www.wikidata.org/wiki/Q745881","display_name":"Intrusion detection system","level":2,"score":0.3215000033378601},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.3086000084877014},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.29429998993873596},{"id":"https://openalex.org/C58489278","wikidata":"https://www.wikidata.org/wiki/Q1172284","display_name":"Data set","level":2,"score":0.28850001096725464},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.2815999984741211},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.267300009727478},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.25920000672340393},{"id":"https://openalex.org/C123201435","wikidata":"https://www.wikidata.org/wiki/Q456632","display_name":"Information privacy","level":2,"score":0.25440001487731934},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.25029999017715454}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.12620","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.12620","pdf_url":"https://arxiv.org/pdf/2609.12620","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2609.12620","doi":"https://doi.org/10.48550/arxiv.2609.12620","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.12620","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.12620","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.12620","pdf_url":"https://arxiv.org/pdf/2609.12620","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7212948709.pdf","grobid_xml":"https://content.openalex.org/works/W7212948709.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4414692030","https://openalex.org/W4414920171","https://openalex.org/W7138273241","https://openalex.org/W4412393932","https://openalex.org/W4393118871","https://openalex.org/W4414920472","https://openalex.org/W4390872775","https://openalex.org/W4400341326","https://openalex.org/W4399198650","https://openalex.org/W4415348291"],"abstract_inverted_index":{"The":[0],"increasing":[1],"adoption":[2],"of":[3,27,185,190],"machine":[4],"learning":[5],"in":[6,64,88,114,145],"network":[7],"and":[8,22,97,118,155,187],"distributed":[9],"security":[10,66],"systems":[11],"has":[12],"created":[13],"an":[14,128],"urgent":[15],"need":[16],"for":[17,62,75],"mechanisms":[18],"that":[19,109],"can":[20],"selectively":[21],"efficiently":[23],"remove":[24],"the":[25,115],"influence":[26,134],"specific":[28],"training":[29,116],"data":[30,36,77,112],"to":[31,47],"eliminate":[32],"compromised":[33],"or":[34],"adversarial":[35],"points":[37,113],"from":[38],"production":[39],"models.":[40],"Privacy":[41],"regulations":[42],"such":[43,90],"as":[44,68,91],"GDPR's":[45],"\\emph{right":[46],"be":[48],"forgotten}":[49],"also":[50],"pose":[51],"similar":[52],"requirements.":[53],"However,":[54],"existing":[55],"approximate":[56],"unlearning":[57,107,136],"techniques":[58],"remain":[59],"computationally":[60,105],"prohibitive":[61],"deployment":[63],"real-world":[65],"systems,":[67,94],"they":[69],"require":[70],"repeated":[71],"expensive":[72],"Hessian-inverse-vector":[73],"computations":[74],"each":[76],"point":[78],"removal,":[79],"creating":[80],"a":[81,104,120,142],"bottleneck":[82],"when":[83],"processing":[84],"multiple":[85],"related":[86],"requests":[87],"scenarios":[89],"intrusion":[92],"detection":[93],"spam":[95],"filters,":[96],"threat":[98],"intelligence":[99],"platforms.":[100],"Thus,":[101],"we":[102],"introduce":[103],"efficient":[106],"framework":[108],"identifies":[110],"correlated":[111],"set":[117],"applies":[119],"theoretically":[121],"derived":[122],"closed-form":[123],"parameter":[124],"update":[125],"rule,":[126],"achieving":[127],"$82\\times$":[129],"wall-clock":[130],"speedup":[131],"over":[132,147],"standard":[133],"function":[135],"while":[137],"preserving":[138],"model":[139],"utility":[140],"with":[141,173,179],"$10^{-2}$":[143],"improvement":[144],"accuracy":[146],"state-of-the-art":[148],"baselines.":[149],"Our":[150,162],"method":[151],"establishes":[152],"theoretical":[153],"guarantees":[154],"ensures":[156],"numerical":[157],"stability":[158],"through":[159],"Hessian":[160],"damping.":[161],"evaluation":[163],"across":[164],"seven":[165],"diverse":[166],"dataset":[167],"architecture":[168],"combinations,":[169],"including":[170],"large-scale":[171],"CIFAR-100":[172],"ResNet-50,":[174],"demonstrates":[175],"superior":[176],"forgetting":[177],"effectiveness,":[178],"membership":[180],"inference":[181],"attack":[182],"success":[183],"rates":[184],"0.660":[186],"tug-of-war":[188],"scores":[189],"0.950.":[191]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-09-15T00:00:00"}