{"id":"https://openalex.org/W7225204182","doi":"https://doi.org/10.48550/arxiv.2611.03445","title":"Error Accumulation in Second-Order Sequential Machine Unlearning","display_name":"Error Accumulation in Second-Order Sequential Machine Unlearning","publication_year":2026,"publication_date":"2026-11-02","ids":{"openalex":"https://openalex.org/W7225204182","doi":"https://doi.org/10.48550/arxiv.2611.03445"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2611.03445","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2611.03445","pdf_url":"https://arxiv.org/pdf/2611.03445","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/2611.03445","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5160543538","display_name":"Chen Shang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I78577930","display_name":"Columbia University","ror":"https://ror.org/00hj8s172","country_code":"US","type":"education","lineage":["https://openalex.org/I78577930"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shang, Chen","raw_affiliation_strings":["Department of Statistics , Columbia University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Statistics , Columbia University","institution_ids":["https://openalex.org/I78577930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011089052","display_name":"Haolin Zou","orcid":"https://orcid.org/0000-0001-9569-433X","observed_orcids":["https://orcid.org/0000-0001-9569-433X"]},"institutions":[{"id":"https://openalex.org/I78577930","display_name":"Columbia University","ror":"https://ror.org/00hj8s172","country_code":"US","type":"education","lineage":["https://openalex.org/I78577930"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zou, Haolin","raw_affiliation_strings":["Department of Statistics , Columbia University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Statistics , Columbia University","institution_ids":["https://openalex.org/I78577930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160540803","display_name":"Binghe Zhu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I78577930","display_name":"Columbia University","ror":"https://ror.org/00hj8s172","country_code":"US","type":"education","lineage":["https://openalex.org/I78577930"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhu, Binghe","raw_affiliation_strings":["Department of Statistics , Columbia University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Statistics , Columbia University","institution_ids":["https://openalex.org/I78577930"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5017771205","display_name":"Arian Maleki","orcid":"https://orcid.org/0000-0002-5626-5206","observed_orcids":["https://orcid.org/0000-0002-5626-5206"]},"institutions":[{"id":"https://openalex.org/I78577930","display_name":"Columbia University","ror":"https://ror.org/00hj8s172","country_code":"US","type":"education","lineage":["https://openalex.org/I78577930"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Maleki, Arian","raw_affiliation_strings":["Department of Electronic Engineering , Columbia University","Department of Statistics , Columbia University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronic Engineering , Columbia University","institution_ids":["https://openalex.org/I78577930"]},{"raw_affiliation_string":"Department of Statistics , Columbia University","institution_ids":["https://openalex.org/I78577930"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I78577930"],"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":0.9980000257492065,"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":0.9980000257492065,"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":0.0013000000035390258,"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/T11045","display_name":"Privacy, Security, and Data Protection","score":9.999999747378752e-05,"subfield":{"id":"https://openalex.org/subfields/3312","display_name":"Sociology and Political Science"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/machine-unlearning","display_name":"machine unlearning","score":0.9900000095367432},{"id":"https://openalex.org/keywords/error-accumulation","display_name":"error accumulation","score":0.972000002861023},{"id":"https://openalex.org/keywords/second-order-optimization","display_name":"second-order optimization","score":0.7429999709129333},{"id":"https://openalex.org/keywords/newton-s-method","display_name":"Newton's method","score":0.7129999995231628},{"id":"https://openalex.org/keywords/conjugate-gradient-method","display_name":"conjugate gradient method","score":0.9919999837875366},{"id":"https://openalex.org/keywords/certification","display_name":"certification","score":0.453000009059906},{"id":"https://openalex.org/keywords/high-dimensional-models","display_name":"high-dimensional models","score":0.5189999938011169},{"id":"https://openalex.org/keywords/population-risk","display_name":"population risk","score":0.6320000290870667},{"id":"https://openalex.org/keywords/ill-conditioned-problems","display_name":"ill-conditioned problems","score":0.7730000019073486}],"concepts":[{"id":"https://openalex.org/C2777984123","wikidata":"https://www.wikidata.org/wiki/Q9248237","display_name":"Pointwise","level":2,"score":0.9559000134468079},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5389000177383423},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5386000275611877},{"id":"https://openalex.org/C2776809875","wikidata":"https://www.wikidata.org/wiki/Q1375963","display_name":"Converse","level":2,"score":0.4893999993801117},{"id":"https://openalex.org/C2778770139","wikidata":"https://www.wikidata.org/wiki/Q1966904","display_name":"Solver","level":2,"score":0.4618000090122223},{"id":"https://openalex.org/C2778029271","wikidata":"https://www.wikidata.org/wiki/Q5421931","display_name":"Extension (predicate logic)","level":2,"score":0.4496999979019165},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.4399999976158142},{"id":"https://openalex.org/C176248197","wikidata":"https://www.wikidata.org/wiki/Q458526","display_name":"Probably approximately correct learning","level":4,"score":0.42890000343322754},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.4090000092983246},{"id":"https://openalex.org/C183356978","wikidata":"https://www.wikidata.org/wiki/Q1779213","display_name":"Tracking error","level":3,"score":0.4049000144004822},{"id":"https://openalex.org/C2775936607","wikidata":"https://www.wikidata.org/wiki/Q466845","display_name":"Tracking (education)","level":2,"score":0.3961000144481659},{"id":"https://openalex.org/C46304622","wikidata":"https://www.wikidata.org/wiki/Q374814","display_name":"Certification","level":2,"score":0.3959999978542328},{"id":"https://openalex.org/C90673727","wikidata":"https://www.wikidata.org/wiki/Q901718","display_name":"Product (mathematics)","level":2,"score":0.3919999897480011},{"id":"https://openalex.org/C207467116","wikidata":"https://www.wikidata.org/wiki/Q4385666","display_name":"Inverse","level":2,"score":0.38440001010894775},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.3806999921798706},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.37610000371932983},{"id":"https://openalex.org/C3770464","wikidata":"https://www.wikidata.org/wiki/Q775963","display_name":"Smoothing","level":2,"score":0.3707999885082245},{"id":"https://openalex.org/C81184566","wikidata":"https://www.wikidata.org/wiki/Q1191895","display_name":"Conjugate gradient method","level":2,"score":0.30799999833106995},{"id":"https://openalex.org/C122383733","wikidata":"https://www.wikidata.org/wiki/Q865920","display_name":"Approximation error","level":2,"score":0.30000001192092896},{"id":"https://openalex.org/C146834321","wikidata":"https://www.wikidata.org/wiki/Q2979672","display_name":"Closure (psychology)","level":2,"score":0.2912999987602234},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2775999903678894}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2611.03445","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2611.03445","pdf_url":"https://arxiv.org/pdf/2611.03445","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.2611.03445","doi":"https://doi.org/10.48550/arxiv.2611.03445","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2611.03445","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2611.03445","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2611.03445","pdf_url":"https://arxiv.org/pdf/2611.03445","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":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7225204182.pdf","grobid_xml":"https://content.openalex.org/works/W7225204182.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7170188192","https://openalex.org/W7221806321","https://openalex.org/W4393929290","https://openalex.org/W4415798088","https://openalex.org/W7221166530","https://openalex.org/W7203725710","https://openalex.org/W4415796690","https://openalex.org/W7214566334","https://openalex.org/W7133237209","https://openalex.org/W7092288443"],"abstract_inverted_index":{"Machine":[0],"unlearning":[1,52],"seeks":[2],"to":[3,46,66,104],"remove":[4],"the":[5,15,73,100,109,123,150],"influence":[6],"of":[7,72,137,149,175],"deleted":[8],"training":[9],"data":[10],"without":[11],"full":[12],"retraining.":[13],"In":[14,30],"sequential":[16,169],"setting,":[17],"deletion":[18],"requests":[19],"arrive":[20],"over":[21,182],"time,":[22],"and":[23,77,86,94,145,160,171],"repeated":[24],"updates":[25,140],"can":[26],"accumulate":[27],"approximation":[28],"errors.":[29],"this":[31],"paper,":[32],"we":[33],"study":[34],"error":[35],"accumulation":[36],"in":[37,128],"high":[38],"dimensions,":[39],"where":[40],"parameter":[41],"count":[42],"$p$":[43],"is":[44],"proportional":[45],"sample":[47],"size":[48],"$n$,":[49],"for":[50,122,142,153,186],"second-order":[51],"methods":[53,177],"under":[54],"two":[55],"privacy":[56],"settings:":[57],"(i)":[58],"\\emph{partial":[59],"unlearning},":[60,80],"which":[61,81],"allows":[62],"unpublished":[63],"internal":[64],"states":[65],"be":[67],"retained":[68],"while":[69],"certified":[70],"versions":[71],"model":[74,107],"are":[75],"released;":[76],"(ii)":[78],"\\emph{perfect":[79],"retains":[82],"only":[83],"published":[84,151],"models":[85],"no":[87],"hidden":[88],"states.":[89],"We":[90,113],"also":[91],"distinguish":[92],"pointwise":[93,154,158],"pathwise":[95,156,161],"certification,":[96,163],"depending":[97],"on":[98],"whether":[99],"certification":[101,144],"guarantee":[102],"applies":[103],"each":[105],"released":[106,111],"or":[108],"whole":[110],"sequence.":[112],"analyze":[114],"conjugate":[115,180],"gradient":[116],"(CG)":[117],"as":[118],"an":[119],"efficient":[120],"solver":[121],"inverse":[124],"Hessian-vector":[125],"product":[126],"(IHVP)":[127],"Newton":[129],"iterations.":[130],"Our":[131],"theory":[132],"shows":[133],"that":[134],"a":[135],"maximum":[136],"1/2/2/3":[138],"Newton--CG":[139],"suffice":[141],"successful":[143],"vanishing":[146],"population-risk":[147],"degradation":[148],"model,":[152],"partial,":[155,157],"perfect":[159,162],"respectively.":[164],"Numerical":[165],"experiments":[166],"demonstrate":[167],"accurate":[168],"tracking":[170],"substantial":[172],"computational":[173],"gains":[174],"Newton-based":[176],"combined":[178],"with":[179],"gradient,":[181],"gradient-based":[183],"updates,":[184],"especially":[185],"ill-conditioned":[187],"problems.":[188]},"counts_by_year":[],"updated_date":"2026-11-06T12:05:46.655746","created_date":"2026-11-04T00:00:00"}