{"id":"https://openalex.org/W4416429438","doi":"https://doi.org/10.1109/tit.2025.3635118","title":"Statistical Inference for Linear Functionals of Online Least-Squares SGD When <i>t</i> \u2273 <i>d</i> <sup>1+\u03b4</sup>","display_name":"Statistical Inference for Linear Functionals of Online Least-Squares SGD When <i>t</i> \u2273 <i>d</i> <sup>1+\u03b4</sup>","publication_year":2025,"publication_date":"2025-11-20","ids":{"openalex":"https://openalex.org/W4416429438","doi":"https://doi.org/10.1109/tit.2025.3635118"},"language":null,"primary_location":{"id":"doi:10.1109/tit.2025.3635118","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tit.2025.3635118","pdf_url":null,"source":{"id":"https://openalex.org/S4502562","display_name":"IEEE Transactions on Information Theory","issn_l":"0018-9448","issn":["0018-9448","1557-9654"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Information Theory","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5031658323","display_name":"Bhavya Agrawalla","orcid":null},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bhavya Agrawalla","raw_affiliation_strings":["Computer Science Department, Carnegie Mellon University, Pittsburgh, PA, USA","Computer Science Department, Carnegie Mellon University, United States"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Computer Science Department, Carnegie Mellon University, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I74973139"]},{"raw_affiliation_string":"Computer Science Department, Carnegie Mellon University, United States","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101496710","display_name":"Krishnakumar Balasubramanian","orcid":"https://orcid.org/0000-0001-5271-9314"},"institutions":[{"id":"https://openalex.org/I84218800","display_name":"University of California, Davis","ror":"https://ror.org/05rrcem69","country_code":"US","type":"education","lineage":["https://openalex.org/I84218800"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Krishnakumar Balasubramanian","raw_affiliation_strings":["Department of Statistics, University of California, Davis, Davis, CA, USA","Department of Statistics, University of California, Davis, United States"],"raw_orcid":"https://orcid.org/0000-0001-5271-9314","affiliations":[{"raw_affiliation_string":"Department of Statistics, University of California, Davis, Davis, CA, USA","institution_ids":["https://openalex.org/I84218800"]},{"raw_affiliation_string":"Department of Statistics, University of California, Davis, United States","institution_ids":["https://openalex.org/I84218800"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5035308611","display_name":"Promit Ghosal","orcid":"https://orcid.org/0000-0003-2378-0006"},"institutions":[{"id":"https://openalex.org/I40347166","display_name":"University of Chicago","ror":"https://ror.org/024mw5h28","country_code":"US","type":"education","lineage":["https://openalex.org/I40347166"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Promit Ghosal","raw_affiliation_strings":["Department of Statistics, The University of Chicago, Chicago, IL, USA","Department of Statistics, University of Chicago, United States"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Statistics, The University of Chicago, Chicago, IL, USA","institution_ids":["https://openalex.org/I40347166"]},{"raw_affiliation_string":"Department of Statistics, University of Chicago, United States","institution_ids":["https://openalex.org/I40347166"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.16996466,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"72","issue":"1","first_page":"447","last_page":"477"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.8967999815940857,"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":0.8967999815940857,"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":0.020899999886751175,"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":0.01850000023841858,"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/inference","display_name":"Inference","score":0.5396000146865845},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.5149000287055969},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.5044999718666077},{"id":"https://openalex.org/keywords/covariance-matrix","display_name":"Covariance matrix","score":0.4887999892234802},{"id":"https://openalex.org/keywords/contrast","display_name":"Contrast (vision)","score":0.48809999227523804},{"id":"https://openalex.org/keywords/central-limit-theorem","display_name":"Central limit theorem","score":0.4740000069141388},{"id":"https://openalex.org/keywords/iterated-function","display_name":"Iterated function","score":0.4368000030517578},{"id":"https://openalex.org/keywords/random-projection","display_name":"Random projection","score":0.43540000915527344},{"id":"https://openalex.org/keywords/projection","display_name":"Projection (relational algebra)","score":0.4296000003814697},{"id":"https://openalex.org/keywords/gaussian-process","display_name":"Gaussian process","score":0.42750000953674316}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5396000146865845},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5162000060081482},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.5149000287055969},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.5044999718666077},{"id":"https://openalex.org/C185142706","wikidata":"https://www.wikidata.org/wiki/Q1134404","display_name":"Covariance matrix","level":2,"score":0.4887999892234802},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.48809999227523804},{"id":"https://openalex.org/C166785042","wikidata":"https://www.wikidata.org/wiki/Q190391","display_name":"Central limit theorem","level":2,"score":0.4740000069141388},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4722000062465668},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.46160000562667847},{"id":"https://openalex.org/C140479938","wikidata":"https://www.wikidata.org/wiki/Q5254619","display_name":"Iterated function","level":2,"score":0.4368000030517578},{"id":"https://openalex.org/C2777036070","wikidata":"https://www.wikidata.org/wiki/Q18393452","display_name":"Random projection","level":2,"score":0.43540000915527344},{"id":"https://openalex.org/C57493831","wikidata":"https://www.wikidata.org/wiki/Q3134666","display_name":"Projection (relational algebra)","level":2,"score":0.4296000003814697},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.42750000953674316},{"id":"https://openalex.org/C205711294","wikidata":"https://www.wikidata.org/wiki/Q176953","display_name":"Rendering (computer graphics)","level":2,"score":0.42559999227523804},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.41659998893737793},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.4108999967575073},{"id":"https://openalex.org/C62644790","wikidata":"https://www.wikidata.org/wiki/Q3454689","display_name":"Variance reduction","level":3,"score":0.3797999918460846},{"id":"https://openalex.org/C134261354","wikidata":"https://www.wikidata.org/wiki/Q938438","display_name":"Statistical inference","level":2,"score":0.3686000108718872},{"id":"https://openalex.org/C122123141","wikidata":"https://www.wikidata.org/wiki/Q176623","display_name":"Random variable","level":2,"score":0.3422999978065491},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.32519999146461487},{"id":"https://openalex.org/C196921405","wikidata":"https://www.wikidata.org/wiki/Q786431","display_name":"Online algorithm","level":2,"score":0.31869998574256897},{"id":"https://openalex.org/C178650346","wikidata":"https://www.wikidata.org/wiki/Q201984","display_name":"Covariance","level":2,"score":0.31690001487731934},{"id":"https://openalex.org/C2777472644","wikidata":"https://www.wikidata.org/wiki/Q16968992","display_name":"Approximate inference","level":3,"score":0.3151000142097473},{"id":"https://openalex.org/C163175372","wikidata":"https://www.wikidata.org/wiki/Q3339222","display_name":"Linear model","level":2,"score":0.30570000410079956},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.29339998960494995},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.28450000286102295},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.27559998631477356},{"id":"https://openalex.org/C64812099","wikidata":"https://www.wikidata.org/wiki/Q176604","display_name":"Random matrix","level":3,"score":0.2718999981880188},{"id":"https://openalex.org/C97256817","wikidata":"https://www.wikidata.org/wiki/Q1462316","display_name":"Spurious relationship","level":2,"score":0.2632000148296356},{"id":"https://openalex.org/C8272713","wikidata":"https://www.wikidata.org/wiki/Q176737","display_name":"Stochastic process","level":2,"score":0.2624000012874603},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.2590999901294708},{"id":"https://openalex.org/C57869625","wikidata":"https://www.wikidata.org/wiki/Q1783502","display_name":"Rate of convergence","level":3,"score":0.2549000084400177},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.2549000084400177},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.25360000133514404},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.2526000142097473}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tit.2025.3635118","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tit.2025.3635118","pdf_url":null,"source":{"id":"https://openalex.org/S4502562","display_name":"IEEE Transactions on Information Theory","issn_l":"0018-9448","issn":["0018-9448","1557-9654"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Information Theory","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W1568229137","https://openalex.org/W1578330309","https://openalex.org/W1965242235","https://openalex.org/W1972711404","https://openalex.org/W1975049860","https://openalex.org/W1994616650","https://openalex.org/W2038190977","https://openalex.org/W2060938961","https://openalex.org/W2066623346","https://openalex.org/W2086161653","https://openalex.org/W2090161364","https://openalex.org/W2127003326","https://openalex.org/W2127855947","https://openalex.org/W2162746441","https://openalex.org/W2525518963","https://openalex.org/W2619334774","https://openalex.org/W2897048230","https://openalex.org/W2963433607","https://openalex.org/W2963490363","https://openalex.org/W2964081784","https://openalex.org/W2966650337","https://openalex.org/W3007705674","https://openalex.org/W3025638325","https://openalex.org/W3043072433","https://openalex.org/W3127970314","https://openalex.org/W3128015591","https://openalex.org/W3165394516","https://openalex.org/W3169728738","https://openalex.org/W3192961449","https://openalex.org/W3195365943","https://openalex.org/W4206742934","https://openalex.org/W4231580960","https://openalex.org/W4313067584","https://openalex.org/W4360879278","https://openalex.org/W4387343669","https://openalex.org/W4389894563","https://openalex.org/W4394842226","https://openalex.org/W4396677240","https://openalex.org/W4403142766","https://openalex.org/W4405731494"],"related_works":[],"abstract_inverted_index":{"Stochastic":[0],"Gradient":[1],"Descent":[2],"(SGD)":[3],"has":[4],"become":[5],"a":[6,45,52,109],"cornerstone":[7],"method":[8],"in":[9,16,51,99,155,171,210,241],"modern":[10],"data":[11],"science.":[12],"However,":[13],"deploying":[14],"SGD":[15,113,239],"high-stakes":[17],"applications":[18],"necessitates":[19],"rigorous":[20],"quantification":[21],"of":[22,39,118,187],"its":[23],"inherent":[24],"uncertainty.":[25],"In":[26,104],"this":[27,221],"work,":[28],"we":[29,106,196],"establish":[30,214],"<italic":[31,53,78,82,122,126,156,165,175,200,215,246,250],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[32,54,79,83,85,123,127,129,157,159,166,168,176,178,180,182,184,201,216,247,251,253],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">non-asymptotic":[33],"Berry\u2013Esseen":[34,91],"bounds</i>":[35,219],"for":[36,62,112,131,205,220,234,238],"linear":[37],"functionals":[38],"online":[40,151,230],"least-squares":[41],"SGD,":[42],"thereby":[43],"providing":[44],"Gaussian":[46],"Central":[47],"Limit":[48],"Theorem":[49],"(CLT)":[50],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">growing-dimensional":[55],"regime</i>.":[56],"Existing":[57],"approaches":[58],"to":[59,88,173],"high-dimensional":[60],"inference":[61],"projection":[63],"parameters,":[64],"such":[65],"as":[66,121],"[1],":[67],"rely":[68],"on":[69],"inverting":[70],"empirical":[71],"covariance":[72],"matrices":[73],"and":[74,97,162,213,231],"require":[75],"at":[76],"least":[77],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">t</i>":[80,124,248],"\u2273":[81,125,249],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">d</i><sup":[84,128,183,252],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">3/2</sup>":[86],"iterations":[87,119],"achieve":[89],"finite-sample":[90],"guarantees,":[92],"rendering":[93],"them":[94],"computationally":[95],"expensive":[96],"restrictive":[98],"the":[100,116,138,174,192,206,211,227,242],"allowable":[101],"dimensional":[102,139],"scaling.":[103],"contrast,":[105],"show":[107],"that":[108],"CLT":[110,212],"holds":[111],"iterates":[114,240],"when":[115],"number":[117],"grows":[120],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">1+\u03b4</sup>":[130],"any":[132],"\u03b4":[133],">":[134],"0,":[135],"significantly":[136],"extending":[137],"regime":[140,245],"permitted":[141],"by":[142],"prior":[143],"works":[144],"while":[145],"improving":[146],"computational":[147],"efficiency.":[148],"The":[149],"proposed":[150],"SGD-based":[152],"procedure":[153],"operates":[154],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">O</i>(<italic":[158,167,177],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">td</i>)":[160],"time":[161],"requires":[163],"only":[164],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">d</i>)":[169],"memory,":[170],"contrast":[172],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">td</i><sup":[179],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">2</sup>+<italic":[181],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">3</sup>)":[185],"runtime":[186],"covariance-inversion":[188],"methods.":[189],"To":[190],"render":[191],"theory":[193],"practically":[194],"applicable,":[195],"further":[197],"develop":[198],"an":[199],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">online":[202],"variance":[203,208],"estimator</i>":[204],"asymptotic":[207],"appearing":[209],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">high-probability":[217],"deviation":[218],"estimator.":[222],"Collectively,":[223],"these":[224],"results":[225],"yield":[226],"first":[228],"fully":[229],"data-driven":[232],"framework":[233],"constructing":[235],"confidence":[236],"intervals":[237],"near-optimal":[243],"scaling":[244],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">1+\u03b4</sup>.":[254]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-11-20T00:00:00"}