{"id":"https://openalex.org/W3198742493","doi":"https://doi.org/10.1109/isit45174.2021.9517811","title":"Self-Regularity of Output Weights for Overparameterized Two-Layer Neural Networks","display_name":"Self-Regularity of Output Weights for Overparameterized Two-Layer Neural Networks","publication_year":2021,"publication_date":"2021-07-12","ids":{"openalex":"https://openalex.org/W3198742493","doi":"https://doi.org/10.1109/isit45174.2021.9517811","mag":"3198742493"},"language":"en","primary_location":{"id":"doi:10.1109/isit45174.2021.9517811","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isit45174.2021.9517811","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE International Symposium on Information Theory (ISIT)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://hdl.handle.net/1721.1/145309","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5031648070","display_name":"David Gamarnik","orcid":null},"institutions":[{"id":"https://openalex.org/I4210109586","display_name":"Moscow Institute of Thermal Technology","ror":"https://ror.org/021es5e59","country_code":"RU","type":"facility","lineage":["https://openalex.org/I4210109586"]}],"countries":["RU"],"is_corresponding":true,"raw_author_name":"David Gamarnik","raw_affiliation_strings":["MIT"],"affiliations":[{"raw_affiliation_string":"MIT","institution_ids":["https://openalex.org/I4210109586"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029207194","display_name":"Eren C. K\u0131z\u0131lda\u011f","orcid":"https://orcid.org/0000-0003-0411-7161"},"institutions":[{"id":"https://openalex.org/I4210109586","display_name":"Moscow Institute of Thermal Technology","ror":"https://ror.org/021es5e59","country_code":"RU","type":"facility","lineage":["https://openalex.org/I4210109586"]}],"countries":["RU"],"is_corresponding":false,"raw_author_name":"Eren C. K\u0131z\u0131lda\u011f","raw_affiliation_strings":["MIT"],"affiliations":[{"raw_affiliation_string":"MIT","institution_ids":["https://openalex.org/I4210109586"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5061028424","display_name":"Ilias Zadik","orcid":"https://orcid.org/0009-0008-4085-324X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ilias Zadik","raw_affiliation_strings":["NYU"],"affiliations":[{"raw_affiliation_string":"NYU","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5031648070"],"corresponding_institution_ids":["https://openalex.org/I4210109586"],"apc_list":null,"apc_paid":null,"fwci":0.1399,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.56165734,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":94},"biblio":{"volume":null,"issue":null,"first_page":"819","last_page":"824"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9998000264167786,"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/T10320","display_name":"Neural Networks and Applications","score":0.9998000264167786,"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/T12676","display_name":"Machine Learning and ELM","score":0.9976999759674072,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9944000244140625,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/artificial-neural-network","display_name":"Artificial neural network","score":0.6778348088264465},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6204644441604614},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.6133729219436646},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.37047070264816284},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3701089024543762},{"id":"https://openalex.org/keywords/materials-science","display_name":"Materials science","score":0.10511711239814758}],"concepts":[{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6778348088264465},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6204644441604614},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.6133729219436646},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.37047070264816284},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3701089024543762},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.10511711239814758},{"id":"https://openalex.org/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/isit45174.2021.9517811","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isit45174.2021.9517811","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE International Symposium on Information Theory (ISIT)","raw_type":"proceedings-article"},{"id":"pmh:oai:dspace.mit.edu:1721.1/145309","is_oa":true,"landing_page_url":"https://hdl.handle.net/1721.1/145309","pdf_url":null,"source":{"id":"https://openalex.org/S4306400425","display_name":"DSpace@MIT (Massachusetts Institute of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I63966007","host_organization_name":"Massachusetts Institute of Technology","host_organization_lineage":["https://openalex.org/I63966007"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Prof. Gamarnik","raw_type":"http://purl.org/eprint/type/ConferencePaper"}],"best_oa_location":{"id":"pmh:oai:dspace.mit.edu:1721.1/145309","is_oa":true,"landing_page_url":"https://hdl.handle.net/1721.1/145309","pdf_url":null,"source":{"id":"https://openalex.org/S4306400425","display_name":"DSpace@MIT (Massachusetts Institute of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I63966007","host_organization_name":"Massachusetts Institute of Technology","host_organization_lineage":["https://openalex.org/I63966007"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Prof. Gamarnik","raw_type":"http://purl.org/eprint/type/ConferencePaper"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":106,"referenced_works":["https://openalex.org/W1720114023","https://openalex.org/W1811750039","https://openalex.org/W1993882792","https://openalex.org/W2059811159","https://openalex.org/W2099579348","https://openalex.org/W2117130368","https://openalex.org/W2131172946","https://openalex.org/W2156387975","https://openalex.org/W2163605009","https://openalex.org/W2169071224","https://openalex.org/W2194775991","https://openalex.org/W2591714514","https://openalex.org/W2596625124","https://openalex.org/W2604117713","https://openalex.org/W2732724430","https://openalex.org/W2736970300","https://openalex.org/W2741952635","https://openalex.org/W2766371994","https://openalex.org/W2767382501","https://openalex.org/W2777138330","https://openalex.org/W2777256551","https://openalex.org/W2785631679","https://openalex.org/W2787248994","https://openalex.org/W2800415562","https://openalex.org/W2808958252","https://openalex.org/W2886067286","https://openalex.org/W2894604724","https://openalex.org/W2899790086","https://openalex.org/W2911867426","https://openalex.org/W2946840143","https://openalex.org/W2949804919","https://openalex.org/W2952318479","https://openalex.org/W2962698540","https://openalex.org/W2962857907","https://openalex.org/W2962930448","https://openalex.org/W2963100491","https://openalex.org/W2963236897","https://openalex.org/W2963239103","https://openalex.org/W2963285844","https://openalex.org/W2963518130","https://openalex.org/W2963664410","https://openalex.org/W2963672337","https://openalex.org/W2963739978","https://openalex.org/W2963794891","https://openalex.org/W2963826371","https://openalex.org/W2964310974","https://openalex.org/W2965497096","https://openalex.org/W2965677819","https://openalex.org/W2970183818","https://openalex.org/W2970332347","https://openalex.org/W2970540478","https://openalex.org/W2971043187","https://openalex.org/W2974929660","https://openalex.org/W2978073653","https://openalex.org/W3013992185","https://openalex.org/W3036841168","https://openalex.org/W3046351808","https://openalex.org/W3093317808","https://openalex.org/W3113714439","https://openalex.org/W3119586787","https://openalex.org/W3134450292","https://openalex.org/W3137695714","https://openalex.org/W4211192810","https://openalex.org/W4252305909","https://openalex.org/W4289436733","https://openalex.org/W4293448676","https://openalex.org/W4297738147","https://openalex.org/W4300588179","https://openalex.org/W6637414576","https://openalex.org/W6638214083","https://openalex.org/W6679168169","https://openalex.org/W6682889407","https://openalex.org/W6684191040","https://openalex.org/W6733941281","https://openalex.org/W6734855675","https://openalex.org/W6735544424","https://openalex.org/W6739659843","https://openalex.org/W6740483536","https://openalex.org/W6741472820","https://openalex.org/W6741653254","https://openalex.org/W6744989570","https://openalex.org/W6745448519","https://openalex.org/W6745751660","https://openalex.org/W6746713790","https://openalex.org/W6747520023","https://openalex.org/W6748256991","https://openalex.org/W6748600614","https://openalex.org/W6748748765","https://openalex.org/W6750606353","https://openalex.org/W6751002639","https://openalex.org/W6752544856","https://openalex.org/W6752591435","https://openalex.org/W6753918066","https://openalex.org/W6755150206","https://openalex.org/W6756001544","https://openalex.org/W6758902322","https://openalex.org/W6761496057","https://openalex.org/W6763368363","https://openalex.org/W6763485134","https://openalex.org/W6767324501","https://openalex.org/W6767695172","https://openalex.org/W6767902025","https://openalex.org/W6775446939","https://openalex.org/W6780065278","https://openalex.org/W6784331473","https://openalex.org/W6791322518"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2051487156","https://openalex.org/W2073681303","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"We":[0,59,158],"consider":[1],"the":[2,33,38,48,51,64,85,119,128,138,142],"problem":[3],"of":[4,50,66,118,121],"finding":[5],"a":[6,22,42,71,92,102,111],"two-layer":[7],"neural":[8],"network":[9,90],"with":[10],"sigmoid,":[11],"rectified":[12],"linear":[13],"unit,":[14],"or":[15],"binary":[16],"step":[17],"activation":[18],"functions":[19],"that":[20,47,78],"\u201cfits\u201d":[21],"training":[23,34,44,94,129],"data":[24,99,139],"set":[25],"as":[26,28,30],"accurately":[27],"possible":[29],"quantified":[31],"by":[32],"error;":[35],"and":[36,131],"study":[37],"following":[39],"question:":[40],"does":[41],"low":[43],"error":[45,95],"guarantee":[46],"norm":[49],"output":[52,68],"layer":[53],"(outer":[54],"norm)":[55],"itself":[56],"is":[57],"small?":[58],"address":[60],"this":[61],"question":[62],"for":[63,171],"case":[65],"non-negative":[67],"weights.":[69],"Using":[70],"simple":[72],"covering":[73],"number":[74,120],"argument,":[75],"we":[76],"establish":[77],"under":[79],"quite":[80,134],"mild":[81,135],"distributional":[82],"assumptions":[83,136],"on":[84,96,137],"input/label":[86],"pairs;":[87],"any":[88],"such":[89,172],"achieving":[91],"small":[93],"polynomially":[97],"many":[98],"necessarily":[100],"has":[101],"well-controlled":[103],"outer":[104],"norm.":[105],"Notably,":[106],"our":[107,162],"results":[108],"(a)":[109],"have":[110,155],"good":[112],"sample":[113],"complexity,":[114],"(b)":[115],"are":[116,125],"independent":[117,156],"hidden":[122],"units,":[123],"(c)":[124],"oblivious":[126],"to":[127,167],"algorithm;":[130],"(d)":[132],"require":[133],"(in":[140],"particular":[141],"input":[143],"vector":[144],"<tex":[145],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[146,151],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$X$</tex>":[147],"\u220a":[148],"R":[149],"<sup":[150],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">d</sup>":[152],"need":[153],"not":[154],"coordinates).":[157],"then":[159],"show":[160],"how":[161],"bounds":[163],"can":[164],"be":[165],"leveraged":[166],"yield":[168],"generalization":[169],"guarantees":[170],"networks.":[173]},"counts_by_year":[{"year":2022,"cited_by_count":1}],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2025-10-10T00:00:00"}