{"id":"https://openalex.org/W4391128820","doi":"https://doi.org/10.1109/tpami.2024.3357172","title":"Implicit Regularization of Dropout","display_name":"Implicit Regularization of Dropout","publication_year":2024,"publication_date":"2024-01-23","ids":{"openalex":"https://openalex.org/W4391128820","doi":"https://doi.org/10.1109/tpami.2024.3357172","pmid":"https://pubmed.ncbi.nlm.nih.gov/38261480"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2024.3357172","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2024.3357172","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","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 Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","pubmed"],"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/A5084678300","display_name":"Zhongwang Zhang","orcid":"https://orcid.org/0009-0006-4202-8556"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Zhongwang Zhang","raw_affiliation_strings":["Institute of Natural Sciences, School of Mathematical Sciences, MOE-LSC and Qing Yuan Research Institute, Shanghai Jiao Tong University, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Institute of Natural Sciences, School of Mathematical Sciences, MOE-LSC and Qing Yuan Research Institute, Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5002236438","display_name":"Zhi\u2010Qin John Xu","orcid":"https://orcid.org/0000-0003-0627-3520"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhi-Qin John Xu","raw_affiliation_strings":["Institute of Natural Sciences, School of Mathematical Sciences, MOE-LSC and Qing Yuan Research Institute, Shanghai Jiao Tong University, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Institute of Natural Sciences, School of Mathematical Sciences, MOE-LSC and Qing Yuan Research Institute, Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5084678300"],"corresponding_institution_ids":["https://openalex.org/I183067930"],"apc_list":null,"apc_paid":null,"fwci":15.1874,"has_fulltext":false,"cited_by_count":43,"citation_normalized_percentile":{"value":0.99205398,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":100},"biblio":{"volume":"46","issue":"6","first_page":"4206","last_page":"4217"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9987999796867371,"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":0.9987999796867371,"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":0.998199999332428,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.987500011920929,"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/regularization","display_name":"Regularization (linguistics)","score":0.7567020058631897},{"id":"https://openalex.org/keywords/dropout","display_name":"Dropout (neural networks)","score":0.6681439280509949},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6216849088668823},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5800914168357849},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5553134679794312},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.5484293103218079},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.5116356015205383},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3546822965145111},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.34947681427001953},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.32435932755470276}],"concepts":[{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.7567020058631897},{"id":"https://openalex.org/C2776145597","wikidata":"https://www.wikidata.org/wiki/Q25339462","display_name":"Dropout (neural networks)","level":2,"score":0.6681439280509949},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6216849088668823},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5800914168357849},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5553134679794312},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.5484293103218079},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.5116356015205383},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3546822965145111},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.34947681427001953},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.32435932755470276}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tpami.2024.3357172","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2024.3357172","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","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 Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},{"id":"pmid:38261480","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/38261480","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE transactions on pattern analysis and machine intelligence","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1339871232","display_name":null,"funder_award_id":"92270001","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G377228980","display_name":null,"funder_award_id":"12371511","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":70,"referenced_works":["https://openalex.org/W1686810756","https://openalex.org/W1746889935","https://openalex.org/W1904365287","https://openalex.org/W2095705004","https://openalex.org/W2142274086","https://openalex.org/W2194775991","https://openalex.org/W2732724430","https://openalex.org/W2777662428","https://openalex.org/W2914483840","https://openalex.org/W2951841718","https://openalex.org/W2963959597","https://openalex.org/W2994720379","https://openalex.org/W3034384909","https://openalex.org/W3038345636","https://openalex.org/W3043000634","https://openalex.org/W3127152051","https://openalex.org/W3137368077","https://openalex.org/W3191617243","https://openalex.org/W3197818647","https://openalex.org/W3212151676","https://openalex.org/W4221158005","https://openalex.org/W4225974998","https://openalex.org/W4298350284","https://openalex.org/W4300971841","https://openalex.org/W4309804652","https://openalex.org/W4360886952","https://openalex.org/W4378501026","https://openalex.org/W4385245566","https://openalex.org/W4393048751","https://openalex.org/W6600213771","https://openalex.org/W6631190155","https://openalex.org/W6636705695","https://openalex.org/W6637373629","https://openalex.org/W6637881880","https://openalex.org/W6640036494","https://openalex.org/W6674330103","https://openalex.org/W6681477337","https://openalex.org/W6682824805","https://openalex.org/W6696879442","https://openalex.org/W6726983090","https://openalex.org/W6740483536","https://openalex.org/W6745259790","https://openalex.org/W6747381837","https://openalex.org/W6748807985","https://openalex.org/W6749036959","https://openalex.org/W6751180888","https://openalex.org/W6751983209","https://openalex.org/W6752242045","https://openalex.org/W6752462968","https://openalex.org/W6752495264","https://openalex.org/W6755044039","https://openalex.org/W6757185729","https://openalex.org/W6762034364","https://openalex.org/W6762718338","https://openalex.org/W6767150047","https://openalex.org/W6772141929","https://openalex.org/W6773718060","https://openalex.org/W6773837390","https://openalex.org/W6779798595","https://openalex.org/W6781378444","https://openalex.org/W6782052167","https://openalex.org/W6783271219","https://openalex.org/W6784468880","https://openalex.org/W6788833324","https://openalex.org/W6797046118","https://openalex.org/W6802776046","https://openalex.org/W6810239458","https://openalex.org/W6839192571","https://openalex.org/W6846519816","https://openalex.org/W6852322563"],"related_works":["https://openalex.org/W2918829236","https://openalex.org/W2895097035","https://openalex.org/W4206903459","https://openalex.org/W2754816816","https://openalex.org/W1998698147","https://openalex.org/W4367591703","https://openalex.org/W4366280654","https://openalex.org/W3160167280","https://openalex.org/W3020853991","https://openalex.org/W4362706668"],"abstract_inverted_index":{"It":[0],"is":[1,37,80,121],"important":[2,169],"to":[3,71,103,124,144],"understand":[4],"how":[5],"dropout,":[6,35],"a":[7,14,27,40,81,108,155],"popular":[8],"regularization":[9,33,120],"method,":[10],"aids":[11],"in":[12,83,136,147],"achieving":[13],"good":[15],"generalization":[16],"solution":[17],"during":[18],"neural":[19,105,149],"network":[20,91,106],"training.":[21],"In":[22],"this":[23],"work,":[24],"we":[25,45,62,95],"present":[26],"theoretical":[28],"derivation":[29],"of":[30,34,42,50,67,158],"an":[31,168],"implicit":[32,52,119],"which":[36,54,88],"validated":[38],"by":[39],"series":[41],"experiments.":[43],"Additionally,":[44],"numerically":[46],"study":[47],"two":[48],"implications":[49],"the":[51,84,90,98,104,118,122,137],"regularization,":[53],"intuitively":[55],"rationalizes":[56],"why":[57],"dropout":[58,101,134,146,159],"helps":[59],"generalization.":[60],"First,":[61],"find":[63,96],"that":[64,97],"input":[65],"weights":[66],"hidden":[68,139],"neurons":[69],"tend":[70],"condense":[72],"on":[73,133],"isolated":[74],"orientations":[75],"trained":[76],"with":[77,100,107,112,161],"dropout.":[78,174],"Condensation":[79],"feature":[82],"non-linear":[85],"learning":[86],"process,":[87],"makes":[89],"less":[92],"complex.":[93],"Second,":[94],"training":[99,148],"leads":[102],"flatter":[109],"minimum":[110],"compared":[111,160],"standard":[113],"gradient":[114,163],"descent":[115,164],"training,":[116],"and":[117,165],"key":[123],"finding":[125],"flat":[126],"solutions.":[127],"Although":[128],"our":[129,141],"theory":[130],"mainly":[131],"focuses":[132],"used":[135],"last":[138],"layer,":[140],"experiments":[142],"apply":[143],"general":[145],"networks.":[150],"This":[151],"work":[152],"points":[153],"out":[154],"distinct":[156],"characteristic":[157],"stochastic":[162],"serves":[166],"as":[167],"basis":[170],"for":[171],"fully":[172],"understanding":[173]},"counts_by_year":[{"year":2026,"cited_by_count":6},{"year":2025,"cited_by_count":31},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":1}],"updated_date":"2026-03-12T08:34:05.389933","created_date":"2025-10-10T00:00:00"}