{"id":"https://openalex.org/W3137585737","doi":"https://doi.org/10.1109/tai.2021.3051357","title":"Can Shallow Neural Networks Beat the Curse of Dimensionality? A Mean Field Training Perspective","display_name":"Can Shallow Neural Networks Beat the Curse of Dimensionality? A Mean Field Training Perspective","publication_year":2020,"publication_date":"2020-10-01","ids":{"openalex":"https://openalex.org/W3137585737","doi":"https://doi.org/10.1109/tai.2021.3051357","mag":"3137585737"},"language":"en","primary_location":{"id":"doi:10.1109/tai.2021.3051357","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tai.2021.3051357","pdf_url":null,"source":{"id":"https://openalex.org/S4210169448","display_name":"IEEE Transactions on Artificial Intelligence","issn_l":"2691-4581","issn":["2691-4581"],"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 Artificial Intelligence","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/A5024215146","display_name":"Stephan Wojtowytsch","orcid":"https://orcid.org/0000-0003-3766-5332"},"institutions":[{"id":"https://openalex.org/I20089843","display_name":"Princeton University","ror":"https://ror.org/00hx57361","country_code":"US","type":"education","lineage":["https://openalex.org/I20089843"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Stephan Wojtowytsch","raw_affiliation_strings":["Program of Applied and Computational Mathematics, Princeton University, Princetion, NJ, USA"],"raw_orcid":"https://orcid.org/0000-0003-3766-5332","affiliations":[{"raw_affiliation_string":"Program of Applied and Computational Mathematics, Princeton University, Princetion, NJ, USA","institution_ids":["https://openalex.org/I20089843"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5071854504","display_name":"E Weinan","orcid":null},"institutions":[{"id":"https://openalex.org/I20089843","display_name":"Princeton University","ror":"https://ror.org/00hx57361","country_code":"US","type":"education","lineage":["https://openalex.org/I20089843"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Weinan E","raw_affiliation_strings":["Department of Mathematics and the Program of Applied and Computational Mathematics, Princeton University, Princetion, NJ, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Mathematics and the Program of Applied and Computational Mathematics, Princeton University, Princetion, NJ, USA","institution_ids":["https://openalex.org/I20089843"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5024215146"],"corresponding_institution_ids":["https://openalex.org/I20089843"],"apc_list":null,"apc_paid":null,"fwci":3.128,"has_fulltext":false,"cited_by_count":41,"citation_normalized_percentile":{"value":0.93329808,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":99},"biblio":{"volume":"1","issue":"2","first_page":"121","last_page":"129"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9997000098228455,"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":0.9997000098228455,"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/T11206","display_name":"Model Reduction and Neural Networks","score":0.9988999962806702,"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"}},{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9986000061035156,"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/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.615624189376831},{"id":"https://openalex.org/keywords/curse-of-dimensionality","display_name":"Curse of dimensionality","score":0.6133080124855042},{"id":"https://openalex.org/keywords/population","display_name":"Population","score":0.5801638960838318},{"id":"https://openalex.org/keywords/lipschitz-continuity","display_name":"Lipschitz continuity","score":0.565708339214325},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5454843044281006},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4740155339241028},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.4480932354927063},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.4314267635345459},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3882913589477539},{"id":"https://openalex.org/keywords/applied-mathematics","display_name":"Applied mathematics","score":0.35640183091163635},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.3503827452659607},{"id":"https://openalex.org/keywords/pure-mathematics","display_name":"Pure mathematics","score":0.2300829291343689}],"concepts":[{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.615624189376831},{"id":"https://openalex.org/C111030470","wikidata":"https://www.wikidata.org/wiki/Q1430460","display_name":"Curse of dimensionality","level":2,"score":0.6133080124855042},{"id":"https://openalex.org/C2908647359","wikidata":"https://www.wikidata.org/wiki/Q2625603","display_name":"Population","level":2,"score":0.5801638960838318},{"id":"https://openalex.org/C22324862","wikidata":"https://www.wikidata.org/wiki/Q652707","display_name":"Lipschitz continuity","level":2,"score":0.565708339214325},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5454843044281006},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4740155339241028},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.4480932354927063},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.4314267635345459},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3882913589477539},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.35640183091163635},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.3503827452659607},{"id":"https://openalex.org/C202444582","wikidata":"https://www.wikidata.org/wiki/Q837863","display_name":"Pure mathematics","level":1,"score":0.2300829291343689},{"id":"https://openalex.org/C149923435","wikidata":"https://www.wikidata.org/wiki/Q37732","display_name":"Demography","level":1,"score":0.0},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tai.2021.3051357","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tai.2021.3051357","pdf_url":null,"source":{"id":"https://openalex.org/S4210169448","display_name":"IEEE Transactions on Artificial Intelligence","issn_l":"2691-4581","issn":["2691-4581"],"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 Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G6388601012","display_name":null,"funder_award_id":"N00014-13-1-0338","funder_id":"https://openalex.org/F4320337345","funder_display_name":"Office of Naval Research"}],"funders":[{"id":"https://openalex.org/F4320337345","display_name":"Office of Naval Research","ror":"https://ror.org/00rk2pe57"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":60,"referenced_works":["https://openalex.org/W385466589","https://openalex.org/W1522579744","https://openalex.org/W1988115241","https://openalex.org/W2103496339","https://openalex.org/W2107822587","https://openalex.org/W2166116275","https://openalex.org/W2173398862","https://openalex.org/W2342070830","https://openalex.org/W2793287646","https://openalex.org/W2796146910","https://openalex.org/W2798986185","https://openalex.org/W2804589149","https://openalex.org/W2894604724","https://openalex.org/W2903586563","https://openalex.org/W2913010492","https://openalex.org/W2938647293","https://openalex.org/W2946052306","https://openalex.org/W2949569332","https://openalex.org/W2952110137","https://openalex.org/W2952204734","https://openalex.org/W2963095610","https://openalex.org/W2963325933","https://openalex.org/W2963759574","https://openalex.org/W2970719462","https://openalex.org/W2992906545","https://openalex.org/W2997997872","https://openalex.org/W3006064982","https://openalex.org/W3010825589","https://openalex.org/W3024317845","https://openalex.org/W3026389946","https://openalex.org/W3034803983","https://openalex.org/W3088285924","https://openalex.org/W3092390915","https://openalex.org/W3099745170","https://openalex.org/W3101985406","https://openalex.org/W3102037514","https://openalex.org/W3104810215","https://openalex.org/W3120785056","https://openalex.org/W4255839052","https://openalex.org/W4287868907","https://openalex.org/W4289115720","https://openalex.org/W4289436733","https://openalex.org/W4297813530","https://openalex.org/W6613206883","https://openalex.org/W6676247742","https://openalex.org/W6681017033","https://openalex.org/W6685630973","https://openalex.org/W6725287276","https://openalex.org/W6749731802","https://openalex.org/W6752009368","https://openalex.org/W6755150206","https://openalex.org/W6756904286","https://openalex.org/W6758728975","https://openalex.org/W6762666122","https://openalex.org/W6764442487","https://openalex.org/W6764632358","https://openalex.org/W6773952175","https://openalex.org/W6783980072","https://openalex.org/W6810952809","https://openalex.org/W7064361409"],"related_works":["https://openalex.org/W4288346661","https://openalex.org/W2947416323","https://openalex.org/W2895097035","https://openalex.org/W4206903459","https://openalex.org/W2754816816","https://openalex.org/W1998698147","https://openalex.org/W3005160514","https://openalex.org/W4366280654","https://openalex.org/W3160167280","https://openalex.org/W3020853991"],"abstract_inverted_index":{"We":[0,80,149,192],"prove":[1],"that":[2,84,155,194],"the":[3,52,76,98,105,112,117,142,167,181,220],"gradient":[4,61,85],"descent":[5,62,86],"training":[6,63,87,176],"of":[7,78,144,219],"a":[8,43,162,214,227],"two-layer":[9,122],"neural":[10,129,189],"network":[11],"on":[12,51,146],"empirical":[13],"or":[14],"population":[15,20],"risk":[16,21],"may":[17,72,139,157],"not":[18,186],"decrease":[19],"at":[22,103],"an":[23,208],"order":[24],"faster":[25],"than":[26],"<inline-formula":[27,53],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[28,54,126],"xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math":[29,55],"notation=\"LaTeX\">$t^{-4/(d-2)}$</tex-math></inline-formula>":[30],"under":[31],"mean":[32,39],"field":[33],"scaling.":[34],"The":[35,175,217],"loss":[36],"functional":[37],"is":[38,222],"squared":[40],"error":[41],"with":[42,88],"Lipschitz-continuous":[44],"target":[45,91,113,163],"function":[46,114,119,164],"and":[47,95,152,185,230],"data":[48,71],"distributed":[49],"uniformly":[50],"notation=\"LaTeX\">$d$</tex-math></inline-formula>":[56],"-dimensional":[57],"unit":[58],"cube.":[59],"Thus":[60],"for":[64,121,213],"fitting":[65],"reasonably":[66],"smooth,":[67],"but":[68,101,138],"truly":[69],"high-dimensional":[70],"be":[73,158,197],"subject":[74],"to":[75,160,188,199,206,223],"curse":[77,143],"dimensionality.":[79],"present":[81],"numerical":[82,153],"evidence":[83,154],"general":[89],"Lipschitz":[90],"functions":[92],"becomes":[93],"slower":[94,96],"as":[97],"dimension":[99],"increases,":[100],"converges":[102],"approximately":[104],"same":[106],"rate":[107],"in":[108,116,133,166,180,204,226],"all":[109],"dimensions":[110],"when":[111],"lies":[115,165],"natural":[118],"space":[120],"ReLU":[123],"networks.":[124,174],"<p":[125],"xmlns:xlink=\"http://www.w3.org/1999/xlink\"><i>Impact":[127],"Statement</i>\u2013Artificial":[128],"networks":[130],"perform":[131],"well":[132],"many":[134],"real":[135],"life":[136],"applications,":[137],"suffer":[140],"from":[141],"dimensionality":[145],"certain":[147],"problems.":[148],"provide":[150],"theoretical":[151],"this":[156,225],"related":[159],"whether":[161],"hypothesis":[168,202],"class":[169],"described":[170],"by":[171],"infinitely":[172],"wide":[173],"dynamics":[177],"are":[178],"considered":[179],"fully":[182],"non-linear":[183],"regime":[184],"reduced":[187],"tangent":[190],"kernels.":[191],"believe":[193],"it":[195],"will":[196],"essential":[198],"study":[200],"these":[201],"classes":[203],"detail":[205],"choose":[207],"appropriate":[209],"machine":[210],"learning":[211],"models":[212],"given":[215],"problem.":[216],"goal":[218],"article":[221],"illustrate":[224],"mathematically":[228],"sound":[229],"numerically":[231],"convincing":[232],"fashion.":[233]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":11},{"year":2024,"cited_by_count":4},{"year":2023,"cited_by_count":7},{"year":2022,"cited_by_count":9},{"year":2021,"cited_by_count":6},{"year":2020,"cited_by_count":1}],"updated_date":"2026-05-14T08:36:36.166977","created_date":"2025-10-10T00:00:00"}