{"id":"https://openalex.org/W3166612201","doi":"https://doi.org/10.1109/tpami.2021.3087328","title":"A Hybrid Stochastic-Deterministic Minibatch Proximal Gradient Method for Efficient Optimization and Generalization","display_name":"A Hybrid Stochastic-Deterministic Minibatch Proximal Gradient Method for Efficient Optimization and Generalization","publication_year":2021,"publication_date":"2021-06-09","ids":{"openalex":"https://openalex.org/W3166612201","doi":"https://doi.org/10.1109/tpami.2021.3087328","mag":"3166612201","pmid":"https://pubmed.ncbi.nlm.nih.gov/34101583"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2021.3087328","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2021.3087328","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":true,"oa_status":"green","oa_url":"https://ink.library.smu.edu.sg/sis_research/8979","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5010883708","display_name":"Pan Zhou","orcid":"https://orcid.org/0000-0003-3400-8943"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pan Zhou","raw_affiliation_strings":["Salesforce, Sea AI Lab of Sea Group, Singapore, Singapore"],"raw_orcid":"https://orcid.org/0000-0003-3400-8943","affiliations":[{"raw_affiliation_string":"Salesforce, Sea AI Lab of Sea Group, Singapore, Singapore","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100776794","display_name":"Xiao\u2013Tong Yuan","orcid":"https://orcid.org/0000-0002-7151-8806"},"institutions":[{"id":"https://openalex.org/I200845125","display_name":"Nanjing University of Information Science and Technology","ror":"https://ror.org/02y0rxk19","country_code":"CN","type":"education","lineage":["https://openalex.org/I200845125"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiao-Tong Yuan","raw_affiliation_strings":["School of Computer and Software, Nanjing University of Information Science and Technology, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-7151-8806","affiliations":[{"raw_affiliation_string":"School of Computer and Software, Nanjing University of Information Science and Technology, Nanjing, China","institution_ids":["https://openalex.org/I200845125"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016399094","display_name":"Zhouchen Lin","orcid":"https://orcid.org/0000-0003-1493-7569"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhouchen Lin","raw_affiliation_strings":["Key Lab. of Machine Perception (MoE), School of EECS, Peking University, Beijing, China","Pazhou Lab, Guangzhou, China"],"raw_orcid":"https://orcid.org/0000-0003-1493-7569","affiliations":[{"raw_affiliation_string":"Key Lab. of Machine Perception (MoE), School of EECS, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"Pazhou Lab, Guangzhou, China","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5074834854","display_name":"Steven C. H. Hoi","orcid":"https://orcid.org/0000-0002-4584-3453"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Steven C.H. Hoi","raw_affiliation_strings":["Salesforce Research, Singapore, Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Salesforce Research, Singapore, Singapore","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.5178,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.58809839,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":96},"biblio":{"volume":"44","issue":"10","first_page":"5933","last_page":"5946"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10500","display_name":"Sparse and Compressive Sensing Techniques","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10500","display_name":"Sparse and Compressive Sensing Techniques","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9987000226974487,"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/T10057","display_name":"Face and Expression Recognition","score":0.9959999918937683,"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/generalization","display_name":"Generalization","score":0.6758943200111389},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5770701766014099},{"id":"https://openalex.org/keywords/stochastic-optimization","display_name":"Stochastic optimization","score":0.5192096829414368},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.49837160110473633},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.47878921031951904},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.43294280767440796},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3599758744239807},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.32227370142936707},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.22665876150131226}],"concepts":[{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.6758943200111389},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5770701766014099},{"id":"https://openalex.org/C194387892","wikidata":"https://www.wikidata.org/wiki/Q1747770","display_name":"Stochastic optimization","level":2,"score":0.5192096829414368},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.49837160110473633},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.47878921031951904},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.43294280767440796},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3599758744239807},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.32227370142936707},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.22665876150131226},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/tpami.2021.3087328","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2021.3087328","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:34101583","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/34101583","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},{"id":"pmh:oai:ink.library.smu.edu.sg:sis_research-9982","is_oa":true,"landing_page_url":"https://ink.library.smu.edu.sg/sis_research/8979","pdf_url":null,"source":{"id":"https://openalex.org/S4306401925","display_name":"Singapore Management University Institutional Knowledge (InK) (Singapore Management University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I79891267","host_organization_name":"Singapore Management University","host_organization_lineage":["https://openalex.org/I79891267"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"https://doi.org/10.1109/TPAMI.2021.3087328","raw_type":"Journal Article"}],"best_oa_location":{"id":"pmh:oai:ink.library.smu.edu.sg:sis_research-9982","is_oa":true,"landing_page_url":"https://ink.library.smu.edu.sg/sis_research/8979","pdf_url":null,"source":{"id":"https://openalex.org/S4306401925","display_name":"Singapore Management University Institutional Knowledge (InK) (Singapore Management University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I79891267","host_organization_name":"Singapore Management University","host_organization_lineage":["https://openalex.org/I79891267"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"https://doi.org/10.1109/TPAMI.2021.3087328","raw_type":"Journal Article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1349274172","display_name":null,"funder_award_id":"2018AAA0100400","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"},{"id":"https://openalex.org/G4280725833","display_name":null,"funder_award_id":"61876090","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G4317995357","display_name":null,"funder_award_id":"61936005","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"},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":52,"referenced_works":["https://openalex.org/W562372631","https://openalex.org/W1523985187","https://openalex.org/W1530699444","https://openalex.org/W1751687266","https://openalex.org/W1967558542","https://openalex.org/W1993962865","https://openalex.org/W1994616650","https://openalex.org/W2101986122","https://openalex.org/W2112269233","https://openalex.org/W2113651538","https://openalex.org/W2118545728","https://openalex.org/W2124541940","https://openalex.org/W2129812935","https://openalex.org/W2135046866","https://openalex.org/W2139338362","https://openalex.org/W2183817801","https://openalex.org/W2738537049","https://openalex.org/W2766400329","https://openalex.org/W2911446756","https://openalex.org/W2963189955","https://openalex.org/W2963607709","https://openalex.org/W2972329375","https://openalex.org/W2989773826","https://openalex.org/W3096659018","https://openalex.org/W3103657382","https://openalex.org/W4236362309","https://openalex.org/W4250658187","https://openalex.org/W4292022450","https://openalex.org/W6603545903","https://openalex.org/W6617009369","https://openalex.org/W6632468910","https://openalex.org/W6638214083","https://openalex.org/W6640396632","https://openalex.org/W6640783478","https://openalex.org/W6676105031","https://openalex.org/W6677553319","https://openalex.org/W6680196509","https://openalex.org/W6683798259","https://openalex.org/W6712439485","https://openalex.org/W6743722771","https://openalex.org/W6747991573","https://openalex.org/W6752314297","https://openalex.org/W6754432982","https://openalex.org/W6754749982","https://openalex.org/W6758323327","https://openalex.org/W6759861439","https://openalex.org/W6763651947","https://openalex.org/W6763749585","https://openalex.org/W6765940612","https://openalex.org/W6767277905","https://openalex.org/W6777926273","https://openalex.org/W6779377023"],"related_works":["https://openalex.org/W4286899070","https://openalex.org/W4323366756","https://openalex.org/W2913271688","https://openalex.org/W4234593354","https://openalex.org/W2893302333","https://openalex.org/W4285259204","https://openalex.org/W4389775782","https://openalex.org/W3207830353","https://openalex.org/W2963926425","https://openalex.org/W3024083111"],"abstract_inverted_index":{"Despite":[0],"the":[1,78,85,161,197,226,232],"success":[2],"of":[3,73,92,145,165,235],"stochastic":[4,14,116],"variance-reduced":[5],"gradient":[6,15,37,117],"(SVRG)":[7],"algorithms":[8,147],"in":[9,203],"solving":[10],"large-scale":[11,65,139],"problems,":[12,140],"their":[13],"complexity":[16,60,142,175],"often":[17],"scales":[18],"linearly":[19],"with":[20,46],"data":[21,151],"size":[22],"and":[23,53,81,131,169,178,188,220],"is":[24,62,122,170,181],"expensive":[25],"for":[26,42,64,128,135,172,176,196],"huge":[27],"data.":[28,210],"Accordingly,":[29],"we":[30,212],"propose":[31],"a":[32,166,206],"hybrid":[33],"stochastic-deterministic":[34],"minibatch":[35,72],"proximal":[36],"(":[38],"<small>HSDMPG</small>":[39,56,96,214,236],")":[40],"algorithm":[41],"strongly":[43,89,217],"convex":[44,90,218],"problems":[45,219],"linear":[47],"prediction":[48],"structure,":[49],"e.g.,":[50],"least":[51],"squares":[52],"logistic/softmax":[54],"regression.":[55],"enjoys":[57],"improved":[58],"computational":[59,233],"that":[61],"data-size-independent":[63],"problems.":[66],"It":[67],"iteratively":[68],"samples":[69],"an":[70,98],"evolving":[71],"individual":[74],"losses":[75,180],"to":[76,215],"estimate":[77],"original":[79],"problem,":[80],"can":[82],"efficiently":[83],"minimize":[84],"sampled":[86],"subproblems.":[87],"For":[88,138],"loss":[91,130],"<inline-formula><tex-math":[93,99,103,120,125,132,155,183,189],"notation=\"LaTeX\">$n$</tex-math></inline-formula>":[94],"components,":[95],"attains":[97],"notation=\"LaTeX\">$\\epsilon$</tex-math></inline-formula>":[100],"-optimization-error":[101],"within":[102],"notation=\"LaTeX\">$\\mathcal":[104,184,190],"{O}":[105,185,191],"\\left(\\kappa":[106],"\\log":[107],"^{\\zeta":[108,113],"+1}\\left(\\frac{1}{\\epsilon":[109],"}\\right)\\frac{1}{\\epsilon":[110],"}\\bigwedge":[111],"n\\log":[112],"}\\left(\\frac{1}{\\epsilon":[114],"}\\right)\\right)$</tex-math></inline-formula>":[115],"evaluations,":[118],"where":[119],"notation=\"LaTeX\">$\\kappa$</tex-math></inline-formula>":[121],"condition":[123],"number,":[124],"notation=\"LaTeX\">$\\zeta":[126,133],"=1$</tex-math></inline-formula>":[127],"quadratic":[129,177],"=2$</tex-math></inline-formula>":[134],"generic":[136,179],"loss.":[137],"our":[141,174],"outperforms":[143],"those":[144],"SVRG-type":[146],"with/without":[148],"dependence":[149],"on":[150],"size.":[152],"Particularly,":[153],"when":[154],"notation=\"LaTeX\">$\\epsilon":[156],"=\\mathcal":[157],"{O}(1/\\sqrt{n})$</tex-math></inline-formula>":[158],"which":[159,195],"matches":[160],"intrinsic":[162],"excess":[163],"error":[164],"learning":[167],"model":[168],"sufficient":[171],"generalization,":[173],"respectively":[182],"(n^{0.5}\\log":[186,192],"^{2}(n))$</tex-math></inline-formula>":[187],"^{3}(n))$</tex-math></inline-formula>":[193],",":[194],"first":[198],"time":[199],"achieves":[200],"optimal":[201],"generalization":[202],"less":[204],"than":[205],"single":[207],"pass":[208],"over":[209,225],"Besides,":[211],"extend":[213],"online":[216],"prove":[221],"its":[222],"higher":[223],"efficiency":[224],"prior":[227],"algorithms.":[228],"Numerical":[229],"results":[230],"demonstrate":[231],"advantages":[234],".":[237]},"counts_by_year":[{"year":2024,"cited_by_count":2},{"year":2021,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}