{"id":"https://openalex.org/W4389472911","doi":"https://doi.org/10.1088/2632-2153/ad13ba","title":"Stochastic gradient descent with random label noises: doubly stochastic models and inference stabilizer","display_name":"Stochastic gradient descent with random label noises: doubly stochastic models and inference stabilizer","publication_year":2023,"publication_date":"2023-12-08","ids":{"openalex":"https://openalex.org/W4389472911","doi":"https://doi.org/10.1088/2632-2153/ad13ba"},"language":"en","primary_location":{"id":"doi:10.1088/2632-2153/ad13ba","is_oa":true,"landing_page_url":"https://doi.org/10.1088/2632-2153/ad13ba","pdf_url":"https://iopscience.iop.org/article/10.1088/2632-2153/ad13ba/pdf","source":{"id":"https://openalex.org/S4210200687","display_name":"Machine Learning Science and Technology","issn_l":"2632-2153","issn":["2632-2153"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320083","host_organization_name":"IOP Publishing","host_organization_lineage":["https://openalex.org/P4310320083","https://openalex.org/P4310311669"],"host_organization_lineage_names":["IOP Publishing","Institute of Physics"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Machine Learning: Science and Technology","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://iopscience.iop.org/article/10.1088/2632-2153/ad13ba/pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5081254155","display_name":"Haoyi Xiong","orcid":"https://orcid.org/0000-0002-5451-3253"},"institutions":[{"id":"https://openalex.org/I98301712","display_name":"Baidu (China)","ror":"https://ror.org/03vs3wt56","country_code":"CN","type":"company","lineage":["https://openalex.org/I98301712"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haoyi Xiong","raw_affiliation_strings":["Baidu Inc, Shangdi Street, Haidian District, Beijing, 100085, CHINA"],"raw_orcid":"https://orcid.org/0000-0002-5451-3253","affiliations":[{"raw_affiliation_string":"Baidu Inc, Shangdi Street, Haidian District, Beijing, 100085, CHINA","institution_ids":["https://openalex.org/I98301712"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100712470","display_name":"Xuhong Li","orcid":"https://orcid.org/0000-0002-2582-8256"},"institutions":[{"id":"https://openalex.org/I98301712","display_name":"Baidu (China)","ror":"https://ror.org/03vs3wt56","country_code":"CN","type":"company","lineage":["https://openalex.org/I98301712"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xuhong Li","raw_affiliation_strings":["Baidu Inc, Shangdi Street, Haidian District, Beijing, 100085, CHINA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Baidu Inc, Shangdi Street, Haidian District, Beijing, 100085, CHINA","institution_ids":["https://openalex.org/I98301712"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020248771","display_name":"Boyang Yu","orcid":"https://orcid.org/0000-0002-1401-9250"},"institutions":[{"id":"https://openalex.org/I98301712","display_name":"Baidu (China)","ror":"https://ror.org/03vs3wt56","country_code":"CN","type":"company","lineage":["https://openalex.org/I98301712"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Boyang Yu","raw_affiliation_strings":["Baidu Inc, Shangdi Street, Haidian District, Beijing, 100085, CHINA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Baidu Inc, Shangdi Street, Haidian District, Beijing, 100085, CHINA","institution_ids":["https://openalex.org/I98301712"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008740867","display_name":"Dongrui Wu","orcid":"https://orcid.org/0000-0002-7153-9703"},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dongrui Wu","raw_affiliation_strings":["Huazhong University of Science and Technology, Hongshan District, Wuhan, 430074, CHINA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology, Hongshan District, Wuhan, 430074, CHINA","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045305860","display_name":"Zhanxing Zhu","orcid":"https://orcid.org/0000-0002-2141-6553"},"institutions":[{"id":"https://openalex.org/I98301712","display_name":"Baidu (China)","ror":"https://ror.org/03vs3wt56","country_code":"CN","type":"company","lineage":["https://openalex.org/I98301712"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhanxing Zhu","raw_affiliation_strings":["Baidu Inc, Changping District, Beijing, 100085, CHINA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Baidu Inc, Changping District, Beijing, 100085, CHINA","institution_ids":["https://openalex.org/I98301712"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5066063885","display_name":"Dejing Dou","orcid":"https://orcid.org/0000-0001-7561-1672"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dejing Dou","raw_affiliation_strings":["BCG, Somwhere, Boston, Massachusetts, 02210, UNITED STATES"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"BCG, Somwhere, Boston, Massachusetts, 02210, UNITED STATES","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":{"value":1600,"currency":"GBP","value_usd":1962},"apc_paid":{"value":1600,"currency":"GBP","value_usd":1962},"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.16793033,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"5","issue":"1","first_page":"015039","last_page":"015039"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12535","display_name":"Machine Learning and Data Classification","score":0.9958999752998352,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.9958999752998352,"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":0.9677000045776367,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.9567000269889832,"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/computer-science","display_name":"Computer science","score":0.5833630561828613},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.5500192642211914},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.5120449662208557},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.4369139075279236},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.43626198172569275},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4069406986236572},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.38758575916290283},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.3669516444206238},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.3557029962539673},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.31077855825424194}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5833630561828613},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.5500192642211914},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.5120449662208557},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.4369139075279236},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.43626198172569275},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4069406986236572},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.38758575916290283},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3669516444206238},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3557029962539673},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.31077855825424194}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1088/2632-2153/ad13ba","is_oa":true,"landing_page_url":"https://doi.org/10.1088/2632-2153/ad13ba","pdf_url":"https://iopscience.iop.org/article/10.1088/2632-2153/ad13ba/pdf","source":{"id":"https://openalex.org/S4210200687","display_name":"Machine Learning Science and Technology","issn_l":"2632-2153","issn":["2632-2153"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320083","host_organization_name":"IOP Publishing","host_organization_lineage":["https://openalex.org/P4310320083","https://openalex.org/P4310311669"],"host_organization_lineage_names":["IOP Publishing","Institute of Physics"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Machine Learning: Science and Technology","raw_type":"journal-article"},{"id":"pmh:oai:eprints.soton.ac.uk:486446","is_oa":true,"landing_page_url":"http://doi.org/10.1088/2632-2153/ad13ba>).","pdf_url":"https://eprints.soton.ac.uk/486446/1/Xiong_et_al_2023_Mach._Learn._Sci._Technol._10.1088_2632_2153_ad13ba.pdf","source":{"id":"https://openalex.org/S4306401019","display_name":"ePrints Soton (University of Southampton)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I43439940","host_organization_name":"University of Southampton","host_organization_lineage":["https://openalex.org/I43439940"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"pmh:oai:doaj.org/article:01e1085ba69a4d02bd8f61206ab99fcb","is_oa":true,"landing_page_url":"https://doaj.org/article/01e1085ba69a4d02bd8f61206ab99fcb","pdf_url":null,"source":{"id":"https://openalex.org/S4306401280","display_name":"DOAJ (DOAJ: Directory of Open Access Journals)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Machine Learning: Science and Technology, Vol 5, Iss 1, p 015039 (2024)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.1088/2632-2153/ad13ba","is_oa":true,"landing_page_url":"https://doi.org/10.1088/2632-2153/ad13ba","pdf_url":"https://iopscience.iop.org/article/10.1088/2632-2153/ad13ba/pdf","source":{"id":"https://openalex.org/S4210200687","display_name":"Machine Learning Science and Technology","issn_l":"2632-2153","issn":["2632-2153"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320083","host_organization_name":"IOP Publishing","host_organization_lineage":["https://openalex.org/P4310320083","https://openalex.org/P4310311669"],"host_organization_lineage_names":["IOP Publishing","Institute of Physics"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Machine Learning: Science and Technology","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4389472911.pdf"},"referenced_works_count":99,"referenced_works":["https://openalex.org/W71165104","https://openalex.org/W1567414581","https://openalex.org/W1821462560","https://openalex.org/W2125930537","https://openalex.org/W2194775991","https://openalex.org/W2335728318","https://openalex.org/W2517692985","https://openalex.org/W2617242334","https://openalex.org/W2618398196","https://openalex.org/W2619334774","https://openalex.org/W2766164908","https://openalex.org/W2768267830","https://openalex.org/W2796146910","https://openalex.org/W2803163563","https://openalex.org/W2809090039","https://openalex.org/W2885593519","https://openalex.org/W2904142648","https://openalex.org/W2905340164","https://openalex.org/W2907762094","https://openalex.org/W2911183980","https://openalex.org/W2912349752","https://openalex.org/W2912497910","https://openalex.org/W2912811302","https://openalex.org/W2919841361","https://openalex.org/W2951841718","https://openalex.org/W2962915600","https://openalex.org/W2963177640","https://openalex.org/W2963433607","https://openalex.org/W2963481221","https://openalex.org/W2963735582","https://openalex.org/W2963959597","https://openalex.org/W2964072432","https://openalex.org/W2981873476","https://openalex.org/W2984169168","https://openalex.org/W2987861506","https://openalex.org/W3005957673","https://openalex.org/W3011558670","https://openalex.org/W3016883099","https://openalex.org/W3028642772","https://openalex.org/W3035160371","https://openalex.org/W3035435178","https://openalex.org/W3036049548","https://openalex.org/W3042609801","https://openalex.org/W3109689953","https://openalex.org/W3118608800","https://openalex.org/W3127183582","https://openalex.org/W3137695714","https://openalex.org/W4287757162","https://openalex.org/W4288324537","https://openalex.org/W4288363504","https://openalex.org/W4288365839","https://openalex.org/W4288623659","https://openalex.org/W4293390340","https://openalex.org/W4299649195","https://openalex.org/W4308734952","https://openalex.org/W4367295640","https://openalex.org/W6638523607","https://openalex.org/W6678818196","https://openalex.org/W6687483927","https://openalex.org/W6691815588","https://openalex.org/W6703116779","https://openalex.org/W6726983090","https://openalex.org/W6738373677","https://openalex.org/W6738534199","https://openalex.org/W6739826692","https://openalex.org/W6741259694","https://openalex.org/W6745410505","https://openalex.org/W6746088618","https://openalex.org/W6751647823","https://openalex.org/W6751757709","https://openalex.org/W6751983209","https://openalex.org/W6752495264","https://openalex.org/W6753772092","https://openalex.org/W6755424776","https://openalex.org/W6756528052","https://openalex.org/W6756977813","https://openalex.org/W6757185729","https://openalex.org/W6758153731","https://openalex.org/W6758338957","https://openalex.org/W6758888417","https://openalex.org/W6758908507","https://openalex.org/W6759134080","https://openalex.org/W6760310830","https://openalex.org/W6762034364","https://openalex.org/W6762242251","https://openalex.org/W6762404521","https://openalex.org/W6763485134","https://openalex.org/W6763749585","https://openalex.org/W6766599388","https://openalex.org/W6770592442","https://openalex.org/W6770809779","https://openalex.org/W6774337062","https://openalex.org/W6775078174","https://openalex.org/W6777926273","https://openalex.org/W6779266730","https://openalex.org/W6780365411","https://openalex.org/W6787972765","https://openalex.org/W6790049375","https://openalex.org/W6843907867"],"related_works":["https://openalex.org/W2055243143","https://openalex.org/W4385609682","https://openalex.org/W4321636575","https://openalex.org/W2357796999","https://openalex.org/W2045526782","https://openalex.org/W2741131631","https://openalex.org/W1986418932","https://openalex.org/W2156919374","https://openalex.org/W35446969","https://openalex.org/W1984019423"],"abstract_inverted_index":{"Abstract":[0],"Random":[1],"label":[2,23,40,55,73,155,248,284,296],"noise":[3,24,56,74,142,249,297],"(or":[4],"observational":[5],"noise)":[6],"widely":[7],"exists":[8],"in":[9,113,325],"practical":[10],"machine":[11],"learning":[12,63],"settings.":[13],"While":[14,97],"previous":[15],"studies":[16,237,261],"primarily":[17],"focused":[18],"on":[19,117,131,199,311],"the":[20,26,35,52,62,68,79,98,108,140,154,224,240,262,275,290,295,320,330],"effects":[21,38,331],"of":[22,28,39,46,65,81,144,181,223,235,243,322,332],"to":[25,33,238,299],"performance":[27],"learning,":[29,268],"our":[30,137,205,229,339],"work":[31,206],"intends":[32],"investigate":[34],"implicit":[36,159,167,241,291,333],"regularization":[37],"noise,":[41,156],"under":[42],"mini-batch":[43,105,151],"sampling":[44,106,152],"settings":[45],"stochastic":[47,85,95],"gradient":[48],"descent":[49],"(SGD),":[50],"with":[51,71,88,221,245,280,302,315],"assumption":[53],"that":[54,174,289],"is":[57,102,148],"unbiased.":[58],"Specifically,":[59],"we":[60,77,231,318],"analyze":[61,239],"dynamics":[64,80],"SGD":[66,82,118,145,210,244],"over":[67,107,153],"quadratic":[69],"loss":[70,110],"unbiased":[72,246,282],"(ULN),":[75],"where":[76,269,317],"model":[78,138,177],"as":[83,112,157,211],"a":[84,93,217],"differentiable":[86],"equation":[87],"two":[89,233],"diffusion":[90,100],"terms":[91],"(namely":[92,183],"doubly":[94],"model).":[96],"first":[99,259],"term":[101,143],"caused":[103,149,293],"by":[104,150,193,294],"(label-noiseless)":[109],"gradients,":[111],"many":[114],"other":[115],"works":[116],"(Zhu":[119],"et":[120,126,195],"al":[121,127,196],"2019":[122],"ICML":[123],"7654\u201363;":[124],"Wu":[125],"2020":[128],"Int.":[129],"Conf.":[130,198],"Machine":[132],"Learning":[133,200],"(PMLR)":[134,202],"pp":[135,203],"10367\u201376),":[136],"investigates":[139],"second":[141],"dynamics,":[146],"which":[147],"an":[158,166,212],"regularizer.":[160],"Our":[161,258,286],"theoretical":[162,340],"analysis":[163],"finds":[164],"such":[165],"regularizer":[168,242,292],"would":[169],"favor":[170],"some":[171],"convergence":[172,222],"points":[173],"could":[175],"stabilize":[176],"outputs":[178,276],"against":[179],"perturbations":[180],"parameters":[182,323],"inference":[184,304],"stability":[185],").":[186],"Though":[187],"similar":[188],"phenomenon":[189],"have":[190],"been":[191],"investigated":[192],"Blanc":[194],"(2020":[197],"Theory":[201],"483\u2013513),":[204],"does":[207],"not":[208],"assume":[209],"Ornstein\u2013Uhlenbeck-like":[213],"process":[214],"and":[215,255,328],"achieves":[216],"more":[218],"generalizable":[219],"result":[220],"approximation":[225],"proved.":[226],"To":[227],"validate":[228],"analysis,":[230],"design":[232],"sets":[234],"empirical":[236],"random":[247,283],"for":[250,266],"deep":[251,267],"neural":[252],"network":[253],"training":[254],"linear":[256,313],"regression.":[257],"experiment":[260,287],"noisy":[263],"self-distillation":[264],"tricks":[265],"student":[270],"networks":[271],"are":[272],"trained":[273],"using":[274],"from":[277],"well-trained":[278],"teachers":[279],"additive":[281],"noise.":[285],"shows":[288],"tends":[298],"select":[300],"models":[301],"improved":[303],"stability.":[305],"We":[306],"also":[307],"carry":[308],"out":[309],"experiments":[310],"SGD-based":[312],"regression":[314],"ULN,":[316],"plot":[319],"trajectories":[321],"learned":[324],"every":[326],"step":[327],"visualize":[329],"regularization.":[334],"The":[335],"results":[336],"back":[337],"up":[338],"findings.":[341]},"counts_by_year":[],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}