{"id":"https://openalex.org/W7161711759","doi":"https://doi.org/10.48550/arxiv.2605.17778","title":"Self-Distillation is Optimal Among Spectral Shrinkage Estimators in Spiked Covariance Models","display_name":"Self-Distillation is Optimal Among Spectral Shrinkage Estimators in Spiked Covariance Models","publication_year":2026,"publication_date":"2026-05-18","ids":{"openalex":"https://openalex.org/W7161711759","doi":"https://doi.org/10.48550/arxiv.2605.17778"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.17778","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.17778","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":null,"license_id":null,"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.2605.17778","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136464782","display_name":"Radu Lecoiu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lecoiu, Radu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085554612","display_name":"Debarghya Mukherjee","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mukherjee, Debarghya","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5001559651","display_name":"Pragya Sur","orcid":"https://orcid.org/0000-0001-8609-843X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sur, Pragya","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"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":0.5052000284194946,"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":0.5052000284194946,"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/T11716","display_name":"Random Matrices and Applications","score":0.07540000230073929,"subfield":{"id":"https://openalex.org/subfields/2613","display_name":"Statistics and Probability"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12072","display_name":"Machine Learning and Algorithms","score":0.04780000075697899,"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/shrinkage","display_name":"Shrinkage","score":0.8278999924659729},{"id":"https://openalex.org/keywords/estimator","display_name":"Estimator","score":0.7809000015258789},{"id":"https://openalex.org/keywords/shrinkage-estimator","display_name":"Shrinkage estimator","score":0.7021999955177307},{"id":"https://openalex.org/keywords/covariance","display_name":"Covariance","score":0.6517000198364258},{"id":"https://openalex.org/keywords/aggregate","display_name":"Aggregate (composite)","score":0.4603999853134155},{"id":"https://openalex.org/keywords/covariance-matrix","display_name":"Covariance matrix","score":0.4377000033855438},{"id":"https://openalex.org/keywords/regression","display_name":"Regression","score":0.3522000014781952},{"id":"https://openalex.org/keywords/estimation-of-covariance-matrices","display_name":"Estimation of covariance matrices","score":0.34610000252723694},{"id":"https://openalex.org/keywords/analysis-of-covariance","display_name":"Analysis of covariance","score":0.32910001277923584}],"concepts":[{"id":"https://openalex.org/C180145272","wikidata":"https://www.wikidata.org/wiki/Q7504144","display_name":"Shrinkage","level":2,"score":0.8278999924659729},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.7809000015258789},{"id":"https://openalex.org/C102592046","wikidata":"https://www.wikidata.org/wiki/Q7504144","display_name":"Shrinkage estimator","level":5,"score":0.7021999955177307},{"id":"https://openalex.org/C178650346","wikidata":"https://www.wikidata.org/wiki/Q201984","display_name":"Covariance","level":2,"score":0.6517000198364258},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.6175000071525574},{"id":"https://openalex.org/C4679612","wikidata":"https://www.wikidata.org/wiki/Q866298","display_name":"Aggregate (composite)","level":2,"score":0.4603999853134155},{"id":"https://openalex.org/C185142706","wikidata":"https://www.wikidata.org/wiki/Q1134404","display_name":"Covariance matrix","level":2,"score":0.4377000033855438},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4246000051498413},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.38989999890327454},{"id":"https://openalex.org/C83546350","wikidata":"https://www.wikidata.org/wiki/Q1139051","display_name":"Regression","level":2,"score":0.3522000014781952},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.34779998660087585},{"id":"https://openalex.org/C180877172","wikidata":"https://www.wikidata.org/wiki/Q5401390","display_name":"Estimation of covariance matrices","level":3,"score":0.34610000252723694},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.33230000734329224},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.33090001344680786},{"id":"https://openalex.org/C119340705","wikidata":"https://www.wikidata.org/wiki/Q1628597","display_name":"Analysis of covariance","level":2,"score":0.32910001277923584},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.31290000677108765},{"id":"https://openalex.org/C207390915","wikidata":"https://www.wikidata.org/wiki/Q1230525","display_name":"Divergence (linguistics)","level":2,"score":0.31209999322891235},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.3018999993801117},{"id":"https://openalex.org/C48921125","wikidata":"https://www.wikidata.org/wiki/Q10861030","display_name":"Linear regression","level":2,"score":0.2971999943256378},{"id":"https://openalex.org/C25343380","wikidata":"https://www.wikidata.org/wiki/Q277521","display_name":"Relation (database)","level":2,"score":0.296999990940094},{"id":"https://openalex.org/C125112378","wikidata":"https://www.wikidata.org/wiki/Q176640","display_name":"Randomness","level":2,"score":0.29660001397132874},{"id":"https://openalex.org/C32277403","wikidata":"https://www.wikidata.org/wiki/Q740445","display_name":"Ridge","level":2,"score":0.2915000021457672},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.2865999937057495},{"id":"https://openalex.org/C184050105","wikidata":"https://www.wikidata.org/wiki/Q273163","display_name":"Isotropy","level":2,"score":0.28299999237060547},{"id":"https://openalex.org/C152877465","wikidata":"https://www.wikidata.org/wiki/Q208042","display_name":"Regression analysis","level":2,"score":0.27900001406669617},{"id":"https://openalex.org/C134261354","wikidata":"https://www.wikidata.org/wiki/Q938438","display_name":"Statistical inference","level":2,"score":0.26420000195503235},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.2639999985694885},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.25870001316070557},{"id":"https://openalex.org/C2986587452","wikidata":"https://www.wikidata.org/wiki/Q938438","display_name":"Statistical analysis","level":2,"score":0.2565000057220459},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.2563000023365021},{"id":"https://openalex.org/C163175372","wikidata":"https://www.wikidata.org/wiki/Q3339222","display_name":"Linear model","level":2,"score":0.250900000333786}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.17778","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.17778","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.17778","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.17778","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Self-distillation":[0],"has":[1],"emerged":[2],"as":[3],"a":[4,31,113,125,165,179],"promising":[5],"technique":[6],"for":[7,43,75,84],"improving":[8],"model":[9],"performance":[10,54,176],"in":[11,23,62],"modern":[12],"machine":[13,65],"learning":[14],"systems.":[15],"We":[16,40,110],"develop":[17],"the":[18,91,142,148,156],"statistical":[19,181],"foundations":[20],"of":[21,34,94,150],"self-distillation":[22,51,173],"spiked":[24,44],"covariance":[25,45],"models,":[26],"by":[27],"introducing":[28],"and":[29,64,124,177],"analyzing":[30],"broad":[32],"class":[33],"estimators,":[35,58],"namely":[36],"spectral":[37,56,107,121],"shrinkage":[38,57,108,122],"estimators.":[39,109],"establish":[41],"that":[42,70,99,141],"matrices":[46],"with":[47,185],"$s$":[48,71],"spikes,":[49],"$s$-step":[50],"achieves":[52],"optimal":[53,134,157],"among":[55,106],"outperforming":[59],"well-known":[60],"estimators":[61,123],"statistics":[63],"learning.":[66],"Moreover,":[67],"we":[68,97,139],"show":[69,98],"steps":[72],"are":[73,161],"necessary":[74],"optimality:":[76],"any":[77],"$(s-k)$-step":[78],"distilled":[79],"estimator":[80],"is":[81],"strictly":[82],"suboptimal":[83],"$1":[85],"\\leq":[86,88],"k":[87],"s$.":[89],"For":[90],"special":[92],"subclass":[93],"isotropic":[95],"covariances,":[96],"optimally":[100],"tuned":[101],"Ridge":[102],"regression":[103],"performs":[104],"best":[105,143],"also":[111],"study":[112],"federated":[114],"approach":[115],"where":[116],"multiple":[117],"data":[118,160],"centers":[119],"share":[120],"common":[126],"server":[127],"seeks":[128],"to":[129,132],"aggregate":[130],"them":[131],"achieve":[133],"performance.":[135],"In":[136],"this":[137],"case,":[138],"find":[140],"local":[144],"rule":[145,158],"again":[146],"takes":[147],"form":[149],"self-distillation,":[151],"though":[152],"it":[153,184],"differs":[154],"from":[155],"when":[159],"hosted":[162],"centrally":[163],"on":[164],"single":[166],"server.":[167],"Together,":[168],"our":[169],"results":[170],"elucidate":[171],"why":[172],"improves":[174],"predictive":[175],"provide":[178],"broader":[180],"framework":[182],"connecting":[183],"classical":[186],"shrinkage-based":[187],"methods.":[188]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-20T00:00:00"}