{"id":"https://openalex.org/W4391533027","doi":"https://doi.org/10.1080/10556788.2024.2306383","title":"Optimized convergence of stochastic gradient descent by weighted averaging","display_name":"Optimized convergence of stochastic gradient descent by weighted averaging","publication_year":2024,"publication_date":"2024-02-05","ids":{"openalex":"https://openalex.org/W4391533027","doi":"https://doi.org/10.1080/10556788.2024.2306383"},"language":"en","primary_location":{"id":"doi:10.1080/10556788.2024.2306383","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1080/10556788.2024.2306383","pdf_url":null,"source":{"id":"https://openalex.org/S103047102","display_name":"Optimization methods & software","issn_l":"1026-7670","issn":["1026-7670","1029-4937","1055-6788"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320547","host_organization_name":"Taylor & Francis","host_organization_lineage":["https://openalex.org/P4310320547","https://openalex.org/P4310320449"],"host_organization_lineage_names":["Taylor & Francis","Informa"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Optimization Methods and Software","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"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/A5000135645","display_name":"Melinda Hagedorn","orcid":"https://orcid.org/0000-0003-2247-7216"},"institutions":[{"id":"https://openalex.org/I44260953","display_name":"Heinrich Heine University D\u00fcsseldorf","ror":"https://ror.org/024z2rq82","country_code":"DE","type":"education","lineage":["https://openalex.org/I44260953"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Melinda Hagedorn","raw_affiliation_strings":["Math.-Nat. Fakult\u00e4t, Heinrich Heine Univ, D\u00fcsseldorf, Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Math.-Nat. Fakult\u00e4t, Heinrich Heine Univ, D\u00fcsseldorf, Germany","institution_ids":["https://openalex.org/I44260953"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5015422470","display_name":"Florian Jarre","orcid":"https://orcid.org/0000-0003-3458-123X"},"institutions":[{"id":"https://openalex.org/I44260953","display_name":"Heinrich Heine University D\u00fcsseldorf","ror":"https://ror.org/024z2rq82","country_code":"DE","type":"education","lineage":["https://openalex.org/I44260953"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Florian Jarre","raw_affiliation_strings":["Math.-Nat. Fakult\u00e4t, Heinrich Heine Univ, D\u00fcsseldorf, Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Math.-Nat. Fakult\u00e4t, Heinrich Heine Univ, D\u00fcsseldorf, Germany","institution_ids":["https://openalex.org/I44260953"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5015422470"],"corresponding_institution_ids":["https://openalex.org/I44260953"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.00968918,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"39","issue":"4","first_page":"699","last_page":"724"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":1.0,"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":1.0,"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/T10500","display_name":"Sparse and Compressive Sensing Techniques","score":0.9993000030517578,"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/T12056","display_name":"Markov Chains and Monte Carlo Methods","score":0.9886000156402588,"subfield":{"id":"https://openalex.org/subfields/2613","display_name":"Statistics and Probability"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/convergence","display_name":"Convergence (economics)","score":0.6047452092170715},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.5882954001426697},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.4645986258983612},{"id":"https://openalex.org/keywords/applied-mathematics","display_name":"Applied mathematics","score":0.43668821454048157},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.31087273359298706},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.13024777173995972},{"id":"https://openalex.org/keywords/economics","display_name":"Economics","score":0.12410163879394531},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.07400166988372803}],"concepts":[{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.6047452092170715},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.5882954001426697},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.4645986258983612},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.43668821454048157},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.31087273359298706},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.13024777173995972},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.12410163879394531},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.07400166988372803},{"id":"https://openalex.org/C50522688","wikidata":"https://www.wikidata.org/wiki/Q189833","display_name":"Economic growth","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1080/10556788.2024.2306383","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1080/10556788.2024.2306383","pdf_url":null,"source":{"id":"https://openalex.org/S103047102","display_name":"Optimization methods & software","issn_l":"1026-7670","issn":["1026-7670","1029-4937","1055-6788"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320547","host_organization_name":"Taylor & Francis","host_organization_lineage":["https://openalex.org/P4310320547","https://openalex.org/P4310320449"],"host_organization_lineage_names":["Taylor & Francis","Informa"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Optimization Methods and Software","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":15,"referenced_works":["https://openalex.org/W242703491","https://openalex.org/W1516903196","https://openalex.org/W1988720110","https://openalex.org/W2007339694","https://openalex.org/W2070493638","https://openalex.org/W2086161653","https://openalex.org/W2162287622","https://openalex.org/W2615070957","https://openalex.org/W2787963935","https://openalex.org/W2964075708","https://openalex.org/W3088745370","https://openalex.org/W3127145884","https://openalex.org/W4226479006","https://openalex.org/W4287330525","https://openalex.org/W4322503680"],"related_works":["https://openalex.org/W2748952813","https://openalex.org/W1979597421","https://openalex.org/W2007980826","https://openalex.org/W4245490552","https://openalex.org/W4225152035","https://openalex.org/W2061531152","https://openalex.org/W3002753104","https://openalex.org/W2077600819","https://openalex.org/W1587224694","https://openalex.org/W2911598644"],"abstract_inverted_index":{"Under":[0],"mild":[1],"assumptions":[2],"stochastic":[3,32,63,126,152,177,215],"gradient":[4,64,216],"methods":[5],"asymptotically":[6],"achieve":[7],"an":[8,23],"optimal":[9,25,45,87,94],"rate":[10],"of":[11,17,31,37,56,61,70,103,120,136,139,175,191,200,225],"convergence":[12],"if":[13],"the":[14,29,34,44,48,54,62,74,80,85,101,125,129,140,176,184,192,201,209,214,231,240],"arithmetic":[15,35,75,161],"mean":[16,36,76],"all":[18,38],"iterates":[19,39,49],"is":[20,72,77,227,234],"returned":[21],"as":[22],"approximate":[24,210],"solution.":[26,88],"However,":[27],"in":[28,53,96,134,218],"absence":[30],"noise,":[33,57],"converges":[40],"considerably":[41],"slower":[42],"to":[43,84,146,160,182,248],"solution":[46,211],"than":[47],"themselves.":[50],"And":[51],"also":[52],"presence":[55],"when":[58,230],"a":[59,67,97,104,117,172,188,205,222,246],"termination":[60,115],"method":[65,217],"after":[66,116],"finite":[68,118],"number":[69,119,224],"steps":[71],"considered,":[73],"not":[78,166],"necessarily":[79],"best":[81],"possible":[82,181],"approximation":[83],"unknown":[86],"This":[89,163,198],"paper":[90],"aims":[91],"at":[92],"identifying":[93],"strategies":[95],"particularly":[98],"simple":[99],"case,":[100],"minimization":[102],"strongly":[105],"convex":[106],"function":[107],"with":[108],"i.":[109,110],"d.":[111],"noise":[112],"terms":[113],"and":[114,128,154,243],"steps.":[121],"Explicit":[122],"formulas":[123],"for":[124],"error":[127,131,153,156,178,194,203,233],"optimality":[130,155,193,202],"are":[132,157],"derived":[133],"dependence":[135],"certain":[137],"parameters":[138,148,185],"SGD":[141],"method.":[142],"The":[143,236],"aim":[144,164],"was":[145,180],"choose":[147],"such":[149,186],"that":[150,187,220,245],"both":[151],"reduced":[158],"compared":[159],"averaging.":[162],"could":[165,195],"be":[167,196,253],"achieved;":[168],"however,":[169],"by":[170,213],"allowing":[171],"slight":[173],"increase":[174],"it":[179],"select":[183],"significant":[189],"reduction":[190,199],"achieved.":[197],"has":[204],"strong":[206],"effect":[207],"on":[208],"generated":[212],"case":[219],"only":[221],"moderate":[223],"iterations":[226],"used":[228],"or":[229],"initial":[232],"large.":[235],"numerical":[237],"examples":[238],"confirm":[239],"theoretical":[241],"results":[242],"suggest":[244],"generalization":[247],"non-quadratic":[249],"objective":[250],"functions":[251],"may":[252],"possible.":[254]},"counts_by_year":[],"updated_date":"2026-08-15T07:11:24.734988","created_date":"2025-10-10T00:00:00"}