{"id":"https://openalex.org/W4410584055","doi":"https://doi.org/10.1109/tpami.2025.3572251","title":"Toward Understanding the Generalizability of Delayed Stochastic Gradient Descent","display_name":"Toward Understanding the Generalizability of Delayed Stochastic Gradient Descent","publication_year":2025,"publication_date":"2025-05-21","ids":{"openalex":"https://openalex.org/W4410584055","doi":"https://doi.org/10.1109/tpami.2025.3572251","pmid":"https://pubmed.ncbi.nlm.nih.gov/40397624"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2025.3572251","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2025.3572251","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/A5019549276","display_name":"Xiaoge Deng","orcid":"https://orcid.org/0000-0003-0622-1202"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaoge Deng","raw_affiliation_strings":["Intelligent Game and Decision Lab, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-0622-1202","affiliations":[{"raw_affiliation_string":"Intelligent Game and Decision Lab, Beijing, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042802004","display_name":"Li Shen","orcid":"https://orcid.org/0000-0001-5659-3464"},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Li Shen","raw_affiliation_strings":["Sun Yat-sen University, Guangzhou, China"],"raw_orcid":"https://orcid.org/0000-0001-5659-3464","affiliations":[{"raw_affiliation_string":"Sun Yat-sen University, Guangzhou, China","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106623069","display_name":"Shengwei Li","orcid":"https://orcid.org/0000-0002-7419-1511"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shengwei Li","raw_affiliation_strings":["College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China","College of Computer Science and Technology, National University of Defense Technology, Hunan, China"],"raw_orcid":"https://orcid.org/0000-0002-7419-1511","affiliations":[{"raw_affiliation_string":"College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China","institution_ids":["https://openalex.org/I170215575"]},{"raw_affiliation_string":"College of Computer Science and Technology, National University of Defense Technology, Hunan, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Tao Sun","orcid":"https://orcid.org/0000-0001-5024-1900"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tao Sun","raw_affiliation_strings":["College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China","College of Computer Science and Technology, National University of Defense Technology, Hunan, China"],"raw_orcid":"https://orcid.org/0000-0001-5024-1900","affiliations":[{"raw_affiliation_string":"College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China","institution_ids":["https://openalex.org/I170215575"]},{"raw_affiliation_string":"College of Computer Science and Technology, National University of Defense Technology, Hunan, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Dongsheng Li","orcid":"https://orcid.org/0000-0001-9743-2034"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dongsheng Li","raw_affiliation_strings":["College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China","College of Computer Science and Technology, National University of Defense Technology, Hunan, China"],"raw_orcid":"https://orcid.org/0000-0001-9743-2034","affiliations":[{"raw_affiliation_string":"College of Computer Science and Technology, National University of Defense Technology, Changsha, Hunan, China","institution_ids":["https://openalex.org/I170215575"]},{"raw_affiliation_string":"College of Computer Science and Technology, National University of Defense Technology, Hunan, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5076879912","display_name":"Dacheng Tao","orcid":null},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]},{"id":"https://openalex.org/I4210103986","display_name":"Jingdong (China)","ror":"https://ror.org/01dkjkq64","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210103986"]}],"countries":["CN","SG"],"is_corresponding":false,"raw_author_name":"Dacheng Tao","raw_affiliation_strings":["JD Explore Academy, Beijing, China","Nanyang Technological University, Singapore"],"raw_orcid":"https://orcid.org/0000-0001-7225-5449","affiliations":[{"raw_affiliation_string":"JD Explore Academy, Beijing, China","institution_ids":["https://openalex.org/I4210103986"]},{"raw_affiliation_string":"Nanyang Technological University, Singapore","institution_ids":["https://openalex.org/I172675005"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"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":0.05010088,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"47","issue":"9","first_page":"7976","last_page":"7986"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9025999903678894,"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":0.9025999903678894,"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/generalizability-theory","display_name":"Generalizability theory","score":0.7752676606178284},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.6756709814071655},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5990772247314453},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.581996500492096},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.47187677025794983},{"id":"https://openalex.org/keywords/descent","display_name":"Descent (aeronautics)","score":0.4216955602169037},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3646184802055359},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3285769820213318},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.26958560943603516},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.18774756789207458},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.1733950972557068},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.07954347133636475}],"concepts":[{"id":"https://openalex.org/C27158222","wikidata":"https://www.wikidata.org/wiki/Q5532422","display_name":"Generalizability theory","level":2,"score":0.7752676606178284},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.6756709814071655},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5990772247314453},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.581996500492096},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.47187677025794983},{"id":"https://openalex.org/C2776637919","wikidata":"https://www.wikidata.org/wiki/Q624380","display_name":"Descent (aeronautics)","level":2,"score":0.4216955602169037},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3646184802055359},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3285769820213318},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.26958560943603516},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.18774756789207458},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.1733950972557068},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.07954347133636475},{"id":"https://openalex.org/C146978453","wikidata":"https://www.wikidata.org/wiki/Q3798668","display_name":"Aerospace engineering","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tpami.2025.3572251","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2025.3572251","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:40397624","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/40397624","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":[{"id":"https://metadata.un.org/sdg/13","score":0.4399999976158142,"display_name":"Climate action"}],"awards":[{"id":"https://openalex.org/G2277931485","display_name":null,"funder_award_id":"62421002","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G3606055544","display_name":null,"funder_award_id":"62025208","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":78,"referenced_works":["https://openalex.org/W1506295250","https://openalex.org/W1551177513","https://openalex.org/W1688975502","https://openalex.org/W1994616650","https://openalex.org/W2049469158","https://openalex.org/W2055186664","https://openalex.org/W2083842231","https://openalex.org/W2108598243","https://openalex.org/W2112796928","https://openalex.org/W2120875792","https://openalex.org/W2148872886","https://openalex.org/W2153635508","https://openalex.org/W2154834860","https://openalex.org/W2162341153","https://openalex.org/W2194775991","https://openalex.org/W2566079294","https://openalex.org/W2944153887","https://openalex.org/W2962952793","https://openalex.org/W2963433607","https://openalex.org/W2991820076","https://openalex.org/W2997269094","https://openalex.org/W3021387869","https://openalex.org/W3091864426","https://openalex.org/W3128307059","https://openalex.org/W3203457432","https://openalex.org/W4382239204","https://openalex.org/W6601076486","https://openalex.org/W6622473587","https://openalex.org/W6638214083","https://openalex.org/W6679563152","https://openalex.org/W6680402377","https://openalex.org/W6684249991","https://openalex.org/W6684859321","https://openalex.org/W6692488971","https://openalex.org/W6692846177","https://openalex.org/W6714512170","https://openalex.org/W6735717691","https://openalex.org/W6738480291","https://openalex.org/W6739554518","https://openalex.org/W6741472820","https://openalex.org/W6745375793","https://openalex.org/W6746900977","https://openalex.org/W6747381837","https://openalex.org/W6751193408","https://openalex.org/W6752495264","https://openalex.org/W6752979356","https://openalex.org/W6754942166","https://openalex.org/W6755150206","https://openalex.org/W6759861439","https://openalex.org/W6767185500","https://openalex.org/W6769302070","https://openalex.org/W6778883912","https://openalex.org/W6778944205","https://openalex.org/W6778993914","https://openalex.org/W6784434538","https://openalex.org/W6785924351","https://openalex.org/W6787606924","https://openalex.org/W6787972765","https://openalex.org/W6790330898","https://openalex.org/W6792091297","https://openalex.org/W6795534559","https://openalex.org/W6797081654","https://openalex.org/W6797570359","https://openalex.org/W6798843436","https://openalex.org/W6803117125","https://openalex.org/W6804108236","https://openalex.org/W6804237116","https://openalex.org/W6810333068","https://openalex.org/W6838579793","https://openalex.org/W6838828081","https://openalex.org/W6838948351","https://openalex.org/W6839164317","https://openalex.org/W6842735619","https://openalex.org/W6844710569","https://openalex.org/W6845438332","https://openalex.org/W6948022635","https://openalex.org/W6981657223","https://openalex.org/W7034108470"],"related_works":["https://openalex.org/W4206903459","https://openalex.org/W2754816816","https://openalex.org/W4366280654","https://openalex.org/W3160167280","https://openalex.org/W4231621013","https://openalex.org/W4362706668","https://openalex.org/W2015288657","https://openalex.org/W3008318776","https://openalex.org/W1977633006","https://openalex.org/W1971945429"],"abstract_inverted_index":{"Stochastic":[0],"gradient":[1,88],"descent":[2],"(SGD)":[3],"performed":[4],"in":[5,13],"an":[6,29],"asynchronous":[7,24,54,70,139],"manner":[8],"plays":[9],"a":[10],"crucial":[11],"role":[12],"training":[14,132],"large-scale":[15],"machine":[16,34],"learning":[17,35],"models.":[18],"However,":[19],"the":[20,51,74,82,86,100,123,129,142,146,156,161],"generalization":[21,42,64,101,143],"performance":[22],"of":[23,85,103,131,145],"delayed":[25,87,147],"SGD,":[26],"which":[27],"is":[28,128],"essential":[30],"metric":[31],"for":[32,67,111],"assessing":[33],"algorithms,":[36],"has":[37],"rarely":[38],"been":[39],"explored.":[40],"Existing":[41],"error":[43,65,102,144],"bounds":[44,98],"are":[45],"rather":[46],"pessimistic":[47],"and":[48,56,108,114,126,160],"cannot":[49],"reveal":[50],"correlation":[52],"between":[53],"delays":[55,140],"generalization.":[57],"In":[58],"this":[59,92],"paper,":[60],"we":[61,79,95],"investigate":[62],"sharper":[63],"bound":[66],"SGD":[68,148],"with":[69],"delay":[71,158],"$\\tau$\u03c4.":[72],"Leveraging":[73],"generating":[75],"function":[76],"analysis":[77,151],"tool,":[78],"first":[80],"establish":[81],"average":[83],"stability":[84],"algorithm.":[89,149],"Based":[90],"on":[91,99],"algorithmic":[93],"stability,":[94],"provide":[96],"upper":[97],"$\\widetilde{\\mathcal":[104,109],"{O}}(\\frac{T-\\tau":[105],"}{n\\tau":[106],"})$O\u02dc(T-\u03c4n\u03c4)":[107],"{O}}(\\frac{1}{n})$O\u02dc(1n)":[110],"quadratic":[112],"convex":[113,116],"strongly":[115],"problems,":[117],"respectively,":[118],"where":[119],"$T$T":[120],"refers":[121],"to":[122,155],"iteration":[124],"number":[125],"$n$n":[127],"amount":[130],"data.":[133],"Our":[134],"theoretical":[135,166],"results":[136,163],"indicate":[137],"that":[138],"reduce":[141],"Analogous":[150],"can":[152],"be":[153],"generalized":[154],"random":[157],"setting,":[159],"experimental":[162],"validate":[164],"our":[165],"findings.":[167]},"counts_by_year":[],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-05-22T00:00:00"}