{"id":"https://openalex.org/W7154510283","doi":"https://doi.org/10.48550/arxiv.2604.12340","title":"Information-Geometric Decomposition of Generalization Error in Unsupervised Learning","display_name":"Information-Geometric Decomposition of Generalization Error in Unsupervised Learning","publication_year":2026,"publication_date":"2026-04-14","ids":{"openalex":"https://openalex.org/W7154510283","doi":"https://doi.org/10.48550/arxiv.2604.12340"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.12340","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.12340","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,"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.2604.12340","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133657317","display_name":"Gilhan Kim","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Kim, Gilhan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5133657317"],"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.29899999499320984,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.29899999499320984,"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/T12814","display_name":"Gaussian Processes and Bayesian Inference","score":0.21539999544620514,"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/T12261","display_name":"Statistical Mechanics and Entropy","score":0.1031000018119812,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.5501000285148621},{"id":"https://openalex.org/keywords/rank","display_name":"Rank (graph theory)","score":0.51419997215271},{"id":"https://openalex.org/keywords/eigenvalues-and-eigenvectors","display_name":"Eigenvalues and eigenvectors","score":0.44429999589920044},{"id":"https://openalex.org/keywords/covariance","display_name":"Covariance","score":0.4221999943256378},{"id":"https://openalex.org/keywords/prior-probability","display_name":"Prior probability","score":0.4075999855995178},{"id":"https://openalex.org/keywords/divergence","display_name":"Divergence (linguistics)","score":0.38769999146461487},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.3856000006198883},{"id":"https://openalex.org/keywords/kurtosis","display_name":"Kurtosis","score":0.3831000030040741},{"id":"https://openalex.org/keywords/principal-component-analysis","display_name":"Principal component analysis","score":0.3788999915122986},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.3774000108242035}],"concepts":[{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.7504000067710876},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5501000285148621},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.51419997215271},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.5016000270843506},{"id":"https://openalex.org/C158693339","wikidata":"https://www.wikidata.org/wiki/Q190524","display_name":"Eigenvalues and eigenvectors","level":2,"score":0.44429999589920044},{"id":"https://openalex.org/C178650346","wikidata":"https://www.wikidata.org/wiki/Q201984","display_name":"Covariance","level":2,"score":0.4221999943256378},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.4075999855995178},{"id":"https://openalex.org/C207390915","wikidata":"https://www.wikidata.org/wiki/Q1230525","display_name":"Divergence (linguistics)","level":2,"score":0.38769999146461487},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.3856000006198883},{"id":"https://openalex.org/C166963901","wikidata":"https://www.wikidata.org/wiki/Q287251","display_name":"Kurtosis","level":2,"score":0.3831000030040741},{"id":"https://openalex.org/C27438332","wikidata":"https://www.wikidata.org/wiki/Q2873","display_name":"Principal component analysis","level":2,"score":0.3788999915122986},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.3774000108242035},{"id":"https://openalex.org/C62354387","wikidata":"https://www.wikidata.org/wiki/Q875399","display_name":"Boundary (topology)","level":2,"score":0.3400000035762787},{"id":"https://openalex.org/C110121322","wikidata":"https://www.wikidata.org/wiki/Q865811","display_name":"Distribution (mathematics)","level":2,"score":0.33469998836517334},{"id":"https://openalex.org/C166785042","wikidata":"https://www.wikidata.org/wiki/Q190391","display_name":"Central limit theorem","level":2,"score":0.3319999873638153},{"id":"https://openalex.org/C4199805","wikidata":"https://www.wikidata.org/wiki/Q2725903","display_name":"Gaussian noise","level":2,"score":0.3260999917984009},{"id":"https://openalex.org/C8038995","wikidata":"https://www.wikidata.org/wiki/Q1152135","display_name":"Unsupervised learning","level":2,"score":0.3160000145435333},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3116999864578247},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.30489999055862427},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.3041999936103821},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.29339998960494995},{"id":"https://openalex.org/C184050105","wikidata":"https://www.wikidata.org/wiki/Q273163","display_name":"Isotropy","level":2,"score":0.2922999858856201},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.2896000146865845},{"id":"https://openalex.org/C171752962","wikidata":"https://www.wikidata.org/wiki/Q255166","display_name":"Kullback\u2013Leibler divergence","level":2,"score":0.28760001063346863},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.2831000089645386},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.28060001134872437},{"id":"https://openalex.org/C42355184","wikidata":"https://www.wikidata.org/wiki/Q1361088","display_name":"Matrix decomposition","level":3,"score":0.28040000796318054},{"id":"https://openalex.org/C61224824","wikidata":"https://www.wikidata.org/wiki/Q2260434","display_name":"Mixture model","level":2,"score":0.2777999937534332},{"id":"https://openalex.org/C149441793","wikidata":"https://www.wikidata.org/wiki/Q200726","display_name":"Probability distribution","level":2,"score":0.27300000190734863},{"id":"https://openalex.org/C2778217198","wikidata":"https://www.wikidata.org/wiki/Q556977","display_name":"Cutoff","level":2,"score":0.2720000147819519},{"id":"https://openalex.org/C98385598","wikidata":"https://www.wikidata.org/wiki/Q1339385","display_name":"Empirical distribution function","level":2,"score":0.27140000462532043},{"id":"https://openalex.org/C177384507","wikidata":"https://www.wikidata.org/wiki/Q1149000","display_name":"Multivariate normal distribution","level":3,"score":0.27129998803138733},{"id":"https://openalex.org/C65778772","wikidata":"https://www.wikidata.org/wiki/Q12345341","display_name":"Asymptotic distribution","level":3,"score":0.2676999866962433},{"id":"https://openalex.org/C122342681","wikidata":"https://www.wikidata.org/wiki/Q330828","display_name":"Skewness","level":2,"score":0.2662000060081482},{"id":"https://openalex.org/C185142706","wikidata":"https://www.wikidata.org/wiki/Q1134404","display_name":"Covariance matrix","level":2,"score":0.2522999942302704}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.12340","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.12340","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,"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.2604.12340","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.12340","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,"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":[],"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],"decompose":[1],"the":[2,8,13,17,51,68,79,111,124,134,140,148,153,182,196],"Kullback--Leibler":[3],"generalization":[4],"error":[5],"(GE)":[6],"--":[7,20,139,151,174,179],"expected":[9],"KL":[10],"divergence":[11],"from":[12,45],"data":[14,30],"distribution":[15],"to":[16,70],"trained":[18],"model":[19,28,41,141],"of":[21,48,123],"unsupervised":[22],"learning":[23],"into":[24],"three":[25],"non-negative":[26],"components:":[27],"error,":[29],"bias,":[31],"and":[32,43,55,87,162,177,186],"variance.":[33],"The":[34,129],"decomposition":[35,125],"is":[36,82,101,195],"exact":[37],"for":[38],"any":[39],"e-flat":[40],"class":[42],"follows":[44],"two":[46],"identities":[47],"information":[49],"geometry:":[50],"generalized":[52],"Pythagorean":[53],"theorem":[54],"a":[56,72,93,107,156,170],"dual":[57],"e-mixture":[58],"variance":[59],"identity.":[60],"As":[61],"an":[62,187],"analytically":[63,188],"tractable":[64],"demonstration,":[65],"we":[66],"apply":[67],"framework":[69],"$\u03b5$-PCA,":[71],"regularized":[73],"principal":[74],"component":[75,122],"analysis":[76],"in":[77],"which":[78,120],"empirical":[80,145],"covariance":[81],"truncated":[83],"at":[84,92],"rank":[85,131],"$N_K$":[86],"discarded":[88],"directions":[89],"are":[90,201],"pinned":[91],"fixed":[94],"noise":[95,149],"floor":[96,150],"$\u03b5$.":[97],"Although":[98],"rank-constrained":[99],"$\u03b5$-PCA":[100],"not":[102],"itself":[103],"e-flat,":[104],"it":[105],"admits":[106],"technical":[108],"reformulation":[109],"with":[110,152],"same":[112],"total":[113],"GE":[114],"on":[115],"isotropic":[116],"Gaussian":[117],"data,":[118],"under":[119],"each":[121],"takes":[126],"closed":[127],"form.":[128],"optimal":[130],"emerges":[132],"as":[133],"cutoff":[135,154],"$\u03bb_{\\mathrm{cut}}^{*}":[136],"=":[137],"\u03b5$":[138],"retains":[142],"exactly":[143],"those":[144],"eigenvalues":[146],"exceeding":[147],"reflecting":[155],"marginal-rate":[157],"balance":[158],"between":[159],"model-error":[160],"gain":[161],"data-bias":[163],"cost.":[164],"A":[165],"boundary":[166],"comparison":[167],"further":[168],"yields":[169],"three-regime":[171],"phase":[172],"diagram":[173],"retain-all,":[175],"interior,":[176],"collapse":[178,190],"separated":[180],"by":[181],"lower":[183],"Marchenko--Pastur":[184],"edge":[185],"computable":[189],"threshold":[191],"$\u03b5_{*}(\u03b1)$,":[192],"where":[193],"$\u03b1$":[194],"dimension-to-sample-size":[197],"ratio.":[198],"All":[199],"claims":[200],"verified":[202],"numerically.":[203]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-04-16T00:00:00"}