{"id":"https://openalex.org/W7203630438","doi":"https://doi.org/10.48550/arxiv.2608.14020","title":"When Does More Correct Data Hurt? Insertion-Stability and the Limits of Dimension-Based Theory","display_name":"When Does More Correct Data Hurt? Insertion-Stability and the Limits of Dimension-Based Theory","publication_year":2026,"publication_date":"2026-08-14","ids":{"openalex":"https://openalex.org/W7203630438","doi":"https://doi.org/10.48550/arxiv.2608.14020"},"language":null,"primary_location":{"id":"pmh:oai:arXiv.org:2608.14020","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.14020","pdf_url":"https://arxiv.org/pdf/2608.14020","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":"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/2608.14020","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5147341716","display_name":"Joseph Sankoorikal Johny","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Johny, Joseph Sankoorikal","raw_affiliation_strings":["Independent Researcher , Seattle , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Independent Researcher , Seattle , USA","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5147341716"],"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/T12072","display_name":"Machine Learning and Algorithms","score":0.9667999744415283,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.9667999744415283,"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":0.008500000461935997,"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":0.0026000000070780516,"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/vc-dimension","display_name":"VC dimension","score":0.7089999914169312},{"id":"https://openalex.org/keywords/littlestone-dimension","display_name":"Littlestone dimension","score":0.9710000157356262},{"id":"https://openalex.org/keywords/pac-learning","display_name":"PAC learning","score":0.527999997138977},{"id":"https://openalex.org/keywords/generalization-error","display_name":"generalization error","score":0.5139999985694885}],"concepts":[{"id":"https://openalex.org/C39927690","wikidata":"https://www.wikidata.org/wiki/Q11197","display_name":"Logarithm","level":2,"score":0.6269999742507935},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.6144999861717224},{"id":"https://openalex.org/C119322782","wikidata":"https://www.wikidata.org/wiki/Q2662236","display_name":"VC dimension","level":2,"score":0.6085000038146973},{"id":"https://openalex.org/C2834757","wikidata":"https://www.wikidata.org/wiki/Q4925424","display_name":"Monotone polygon","level":2,"score":0.5967000126838684},{"id":"https://openalex.org/C146834321","wikidata":"https://www.wikidata.org/wiki/Q2979672","display_name":"Closure (psychology)","level":2,"score":0.5863000154495239},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.5735999941825867},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5149999856948853},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.5005999803543091},{"id":"https://openalex.org/C118615104","wikidata":"https://www.wikidata.org/wiki/Q121416","display_name":"Discrete mathematics","level":1,"score":0.4952999949455261},{"id":"https://openalex.org/C136815107","wikidata":"https://www.wikidata.org/wiki/Q154242","display_name":"Nothing","level":2,"score":0.36070001125335693},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.31869998574256897},{"id":"https://openalex.org/C129848803","wikidata":"https://www.wikidata.org/wiki/Q2564360","display_name":"Sample size determination","level":2,"score":0.29260000586509705},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.2874999940395355},{"id":"https://openalex.org/C112313634","wikidata":"https://www.wikidata.org/wiki/Q7886648","display_name":"Complement (music)","level":5,"score":0.2736999988555908},{"id":"https://openalex.org/C102944935","wikidata":"https://www.wikidata.org/wiki/Q4953701","display_name":"Bracketing (phenomenology)","level":2,"score":0.2678999900817871},{"id":"https://openalex.org/C2776299755","wikidata":"https://www.wikidata.org/wiki/Q432449","display_name":"Carry (investment)","level":2,"score":0.266400009393692},{"id":"https://openalex.org/C2777998813","wikidata":"https://www.wikidata.org/wiki/Q16869124","display_name":"Append","level":2,"score":0.2572999894618988},{"id":"https://openalex.org/C774472","wikidata":"https://www.wikidata.org/wiki/Q6760393","display_name":"Margin (machine learning)","level":2,"score":0.251800000667572},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.25119999051094055}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.14020","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.14020","pdf_url":"https://arxiv.org/pdf/2608.14020","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.14020","doi":"https://doi.org/10.48550/arxiv.2608.14020","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.14020","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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.14020","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.14020","pdf_url":"https://arxiv.org/pdf/2608.14020","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7203630438.pdf","grobid_xml":"https://content.openalex.org/works/W7203630438.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Adding":[0],"data":[1],"known":[2],"to":[3,7,130,295],"be":[4,8],"correct":[5,285],"ought":[6],"safe.":[9],"Not":[10],"always.":[11],"Larsen,":[12],"Pabbaraju":[13],"and":[14,29,99,155,167,202,208,219,289],"Shetty":[15],"model":[16],"the":[17,40,50,75,97,100,104,121,131,138,144,147,187,209,231,249,270,292,296,300],"failure":[18,232],"with":[19],"a":[20,71,83,108,234,258,264],"monotone":[21,240],"adversary,":[22,132],"which":[23,93],"reads":[24],"an":[25,276],"i.i.d.":[26],"training":[27],"sample":[28,137],"may":[30],"append":[31],"as":[32,36],"many":[33],"further":[34],"examples":[35,117],"it":[37,89,113,124,158],"likes,":[38],"provided":[39],"target":[41],"hypothesis":[42],"labels":[43],"them":[44],"all.":[45],"Mehrotra":[46],"has":[47],"since":[48,133],"settled":[49],"cost,":[51],"showing":[52,237],"that":[53,80,238,279],"for":[54],"classes":[55,94,193],"of":[56,179,233,244],"VC":[57],"dimension":[58,218],"d":[59],">=":[60],"2":[61,201],"no":[62,239],"learner":[63,109,265,278,301],"can":[64,118,190,194],"guarantee":[65],"expected":[66],"error":[67],"better":[68],"than":[69,230],"Theta((d/n)log(en/d)),":[70],"logarithmic":[72],"factor":[73],"above":[74],"clean":[76,148,177,250,283],"PAC":[77],"rate.":[78,251],"Because":[79],"rate":[81,178],"is":[82,153,159,170,184,254,260,266,280],"worst":[84],"case":[85],"over":[86,165],"all":[87],"classes,":[88],"says":[90],"nothing":[91],"about":[92],"actually":[95],"suffer":[96],"penalty,":[98],"answer":[101],"turns":[102],"on":[103,134,146,196,282],"learner.":[105],"We":[106],"call":[107],"insertion-stable":[110,171,277],"if":[111],"feeding":[112],"more":[114,229],"correctly":[115],"labeled":[116],"only":[119],"shrink":[120],"region":[122],"where":[123],"errs.":[125],"Such":[126],"learners":[127],"are":[128,220,287],"immune":[129,221],"any":[135,245],"given":[136],"risk":[139,145],"after":[140],"insertions":[141],"never":[142],"exceeds":[143],"part":[149],"alone,":[150],"however":[151,156],"much":[152],"added":[154],"cleverly":[157],"chosen.":[160],"High-":[161],"probability":[162],"guarantees":[163],"carry":[164],"unchanged,":[166],"because":[168],"Closure":[169],"every":[172],"intersection-closed":[173],"class":[174,226,259],"keeps":[175],"its":[176],"E[Err]":[180],"<=":[181],"(21d+34)/n.":[182],"Immunity":[183],"not":[185,256,303],"something":[186],"classical":[188],"dimensions":[189],"predict.":[191],"Two":[192],"agree":[195],"VCdim":[197],"=":[198,200],"Ldim":[199],"still":[203],"split,":[204],"one":[205,291],"at":[206,211],"Theta(1/n)":[207],"other":[210],"Theta(log(en)/n),":[212],"while":[213],"intervals":[214],"have":[215],"unbounded":[216],"Littlestone":[217],"anyway.":[222],"On":[223],"Mehrotra's":[224],"hard":[225],"we":[227],"prove":[228],"single":[235],"algorithm,":[236],"permutation-invariant":[241],"compression":[242],"scheme":[243],"finite":[246],"size":[247],"attains":[248],"The":[252],"question":[253],"therefore":[255],"whether":[257,263,269],"hard,":[261],"nor":[262],"good,":[267],"but":[268],"two":[271],"suit":[272],"each":[273],"other.":[274],"Given":[275],"optimal":[281],"data,":[284],"additions":[286],"free,":[288],"without":[290],"cost":[293],"belongs":[294],"class,":[297],"so":[298],"changing":[299],"will":[302],"avoid":[304],"it.":[305]},"counts_by_year":[],"updated_date":"2026-09-30T09:27:27.532063","created_date":"2026-08-18T00:00:00"}