{"id":"https://openalex.org/W4303446428","doi":"https://doi.org/10.1007/s10994-022-06243-3","title":"Scaling up stochastic gradient descent for non-convex optimisation","display_name":"Scaling up stochastic gradient descent for non-convex optimisation","publication_year":2022,"publication_date":"2022-10-07","ids":{"openalex":"https://openalex.org/W4303446428","doi":"https://doi.org/10.1007/s10994-022-06243-3"},"language":"en","primary_location":{"id":"doi:10.1007/s10994-022-06243-3","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s10994-022-06243-3","pdf_url":"https://link.springer.com/content/pdf/10.1007/s10994-022-06243-3.pdf","source":{"id":"https://openalex.org/S62148650","display_name":"Machine Learning","issn_l":"0885-6125","issn":["0885-6125","1573-0565"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"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","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://link.springer.com/content/pdf/10.1007/s10994-022-06243-3.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5103229131","display_name":"Saad Mohamad","orcid":null},"institutions":[{"id":"https://openalex.org/I9300472","display_name":"Bournemouth University","ror":"https://ror.org/05wwcw481","country_code":"GB","type":"education","lineage":["https://openalex.org/I9300472"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Saad Mohamad","raw_affiliation_strings":["Department of Computing, Bournemouth University, Poole, UK"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computing, Bournemouth University, Poole, UK","institution_ids":["https://openalex.org/I9300472"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5009099773","display_name":"Hamad Alamri","orcid":null},"institutions":[{"id":"https://openalex.org/I39555362","display_name":"University of Warwick","ror":"https://ror.org/01a77tt86","country_code":"GB","type":"education","lineage":["https://openalex.org/I39555362"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Hamad Alamri","raw_affiliation_strings":["WMG, Warwick University, Coventry, UK"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"WMG, Warwick University, Coventry, UK","institution_ids":["https://openalex.org/I39555362"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5063662012","display_name":"Abdelhamid Bouchachia","orcid":"https://orcid.org/0000-0002-1980-5517"},"institutions":[{"id":"https://openalex.org/I9300472","display_name":"Bournemouth University","ror":"https://ror.org/05wwcw481","country_code":"GB","type":"education","lineage":["https://openalex.org/I9300472"]}],"countries":["GB"],"is_corresponding":true,"raw_author_name":"Abdelhamid Bouchachia","raw_affiliation_strings":["Department of Computing, Bournemouth University, Poole, UK"],"raw_orcid":"https://orcid.org/0000-0002-1980-5517","affiliations":[{"raw_affiliation_string":"Department of Computing, Bournemouth University, Poole, UK","institution_ids":["https://openalex.org/I9300472"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5063662012"],"corresponding_institution_ids":["https://openalex.org/I9300472"],"apc_list":{"value":3290,"currency":"USD","value_usd":3290},"apc_paid":{"value":3290,"currency":"USD","value_usd":3290},"fwci":0.4002,"has_fulltext":true,"cited_by_count":5,"citation_normalized_percentile":{"value":0.69855457,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":98},"biblio":{"volume":"111","issue":"11","first_page":"4039","last_page":"4079"},"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":0.9980999827384949,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9919000267982483,"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/algorithm","display_name":"Algorithm","score":0.6614595651626587},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6390016078948975},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.6206216216087341},{"id":"https://openalex.org/keywords/convex-function","display_name":"Convex function","score":0.4621947407722473},{"id":"https://openalex.org/keywords/bounded-function","display_name":"Bounded function","score":0.44641175866127014},{"id":"https://openalex.org/keywords/regular-polygon","display_name":"Regular polygon","score":0.36286646127700806},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.31597626209259033},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.21790581941604614},{"id":"https://openalex.org/keywords/geometry","display_name":"Geometry","score":0.1306033432483673},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.09699472784996033},{"id":"https://openalex.org/keywords/mathematical-analysis","display_name":"Mathematical analysis","score":0.07150256633758545}],"concepts":[{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.6614595651626587},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6390016078948975},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.6206216216087341},{"id":"https://openalex.org/C145446738","wikidata":"https://www.wikidata.org/wiki/Q319913","display_name":"Convex function","level":3,"score":0.4621947407722473},{"id":"https://openalex.org/C34388435","wikidata":"https://www.wikidata.org/wiki/Q2267362","display_name":"Bounded function","level":2,"score":0.44641175866127014},{"id":"https://openalex.org/C112680207","wikidata":"https://www.wikidata.org/wiki/Q714886","display_name":"Regular polygon","level":2,"score":0.36286646127700806},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.31597626209259033},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.21790581941604614},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.1306033432483673},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.09699472784996033},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.07150256633758545}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1007/s10994-022-06243-3","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s10994-022-06243-3","pdf_url":"https://link.springer.com/content/pdf/10.1007/s10994-022-06243-3.pdf","source":{"id":"https://openalex.org/S62148650","display_name":"Machine Learning","issn_l":"0885-6125","issn":["0885-6125","1573-0565"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"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","raw_type":"journal-article"},{"id":"pmh:oai:eprints.bournemouth.ac.uk:37680","is_oa":false,"landing_page_url":"https://eprints.bournemouth.ac.uk/37680/7/s10994-022-06243-3.pdf","pdf_url":"https://eprints.bournemouth.ac.uk/37680/7/s10994-022-06243-3.pdf","source":{"id":"https://openalex.org/S4306400187","display_name":"Bournemouth University Research Online (Bournemouth University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I9300472","host_organization_name":"Bournemouth University","host_organization_lineage":["https://openalex.org/I9300472"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"","raw_type":"Article"},{"id":"pmh:oai:wrap.warwick.ac.uk:170693","is_oa":false,"landing_page_url":"https://wrap.warwick.ac.uk/170693/1/WRAP-Scaling-up-stochastic-gradient-descent-for-non-convex-optimisation-Alamri-2022.pdf","pdf_url":"https://wrap.warwick.ac.uk/170693/1/WRAP-Scaling-up-stochastic-gradient-descent-for-non-convex-optimisation-Alamri-2022.pdf","source":{"id":"https://openalex.org/S4306400665","display_name":"Warwick Research Archive Portal (University of Warwick)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I39555362","host_organization_name":"University of Warwick","host_organization_lineage":["https://openalex.org/I39555362"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"","raw_type":"Journal Article"}],"best_oa_location":{"id":"doi:10.1007/s10994-022-06243-3","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s10994-022-06243-3","pdf_url":"https://link.springer.com/content/pdf/10.1007/s10994-022-06243-3.pdf","source":{"id":"https://openalex.org/S62148650","display_name":"Machine Learning","issn_l":"0885-6125","issn":["0885-6125","1573-0565"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"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","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.41999998688697815,"id":"https://metadata.un.org/sdg/8","display_name":"Decent work and economic growth"}],"awards":[{"id":"https://openalex.org/G3068212542","display_name":null,"funder_award_id":"687691","funder_id":"https://openalex.org/F4320332999","funder_display_name":"Horizon 2020 Framework Programme"}],"funders":[{"id":"https://openalex.org/F4320320279","display_name":"University of Warwick","ror":"https://ror.org/01a77tt86"},{"id":"https://openalex.org/F4320332999","display_name":"Horizon 2020 Framework Programme","ror":"https://ror.org/00k4n6c32"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4303446428.pdf","grobid_xml":"https://content.openalex.org/works/W4303446428.grobid-xml"},"referenced_works_count":60,"referenced_works":["https://openalex.org/W114517082","https://openalex.org/W809736386","https://openalex.org/W1095970959","https://openalex.org/W1442374986","https://openalex.org/W1516111018","https://openalex.org/W1842246424","https://openalex.org/W1927459353","https://openalex.org/W1992208280","https://openalex.org/W1994616650","https://openalex.org/W2049469158","https://openalex.org/W2060393849","https://openalex.org/W2072566913","https://openalex.org/W2100787464","https://openalex.org/W2119391823","https://openalex.org/W2120340025","https://openalex.org/W2130062883","https://openalex.org/W2138243089","https://openalex.org/W2145339207","https://openalex.org/W2150468603","https://openalex.org/W2154834860","https://openalex.org/W2155027007","https://openalex.org/W2163605009","https://openalex.org/W2169209873","https://openalex.org/W2225156818","https://openalex.org/W2257979135","https://openalex.org/W2260756217","https://openalex.org/W2492794003","https://openalex.org/W2513289469","https://openalex.org/W2559655401","https://openalex.org/W2561261090","https://openalex.org/W2604421180","https://openalex.org/W2619833911","https://openalex.org/W2744599870","https://openalex.org/W2750933313","https://openalex.org/W2765170276","https://openalex.org/W2898280316","https://openalex.org/W2900182564","https://openalex.org/W2963139509","https://openalex.org/W2963433607","https://openalex.org/W2963468355","https://openalex.org/W2963470657","https://openalex.org/W2963611890","https://openalex.org/W2963903325","https://openalex.org/W2982191236","https://openalex.org/W3038057116","https://openalex.org/W3102816259","https://openalex.org/W3102856533","https://openalex.org/W3127411449","https://openalex.org/W4214717370","https://openalex.org/W4230563027","https://openalex.org/W6638803421","https://openalex.org/W6639619044","https://openalex.org/W6677916085","https://openalex.org/W6684249991","https://openalex.org/W6684578138","https://openalex.org/W6684859321","https://openalex.org/W6713134421","https://openalex.org/W6718047147","https://openalex.org/W6756040250","https://openalex.org/W6763775535"],"related_works":["https://openalex.org/W2051487156","https://openalex.org/W2073681303","https://openalex.org/W3013085049","https://openalex.org/W2544423928","https://openalex.org/W2052122378","https://openalex.org/W2053286651","https://openalex.org/W1560496689","https://openalex.org/W1793337513","https://openalex.org/W2296389180","https://openalex.org/W4387635768"],"abstract_inverted_index":{"Abstract":[0],"Stochastic":[1],"gradient":[2],"descent":[3],"(SGD)":[4],"is":[5,54,99,166,176],"a":[6,23,71,95,103,199,209],"widely":[7],"adopted":[8],"iterative":[9],"method":[10],"for":[11,118],"optimizing":[12],"differentiable":[13],"objective":[14],"functions.":[15],"In":[16],"this":[17],"paper,":[18],"we":[19],"propose":[20,70],"and":[21,35,48,60,74,87,109,128,147,171,207,225],"discuss":[22],"novel":[24],"approach":[25],"to":[26,101,138,142,237],"scale":[27],"up":[28],"SGD":[29,78],"in":[30,125,221],"applications":[31],"involving":[32],"non-convex":[33,119],"functions":[34],"large":[36],"datasets.":[37],"We":[38,69],"address":[39],"the":[40,52,63,143,162,172,182,239,242,246],"bottleneck":[41],"problem":[42,203,214],"arising":[43,124],"when":[44],"using":[45],"both":[46,84],"shared":[47],"distributed":[49,73],"memory.":[50],"Typically,":[51],"former":[53],"bounded":[55,167,177],"by":[56,168,178,193],"limited":[57],"computation":[58,108],"resources":[59],"bandwidth":[61],"whereas":[62],"latter":[64],"suffers":[65],"from":[66],"communication":[67],"overheads.":[68],"unified":[72,96],"parallel":[75],"implementation":[76],"of":[77,114,145,149,157,164,174,184,241],"(named":[79],"DPSGD)":[80],"that":[81,135,161,189],"relies":[82],"on":[83,198,208],"asynchronous":[85],"distribution":[86],"lock-free":[88],"parallelism.":[89],"By":[90],"combining":[91],"two":[92,222],"strategies":[93],"into":[94],"framework,":[97],"DPSGD":[98,115,136,194,244],"able":[100],"strike":[102],"better":[104],"trade-off":[105],"between":[106],"local":[107],"communication.":[110],"The":[111,186],"convergence":[112,155],"properties":[113],"are":[116,195,235],"studied":[117],"problems":[120],"such":[121],"as":[122],"those":[123],"statistical":[126],"modelling":[127],"machine":[129],"learning.":[130],"Our":[131],"theoretical":[132,231],"analysis":[133],"shows":[134],"leads":[137],"speed-up":[139],"with":[140],"respect":[141],"number":[144,148,163,173,183],"cores":[146,165],"workers":[150,175],"while":[151],"guaranteeing":[152],"an":[153],"asymptotic":[154],"rate":[156],"$$O(1/\\sqrt{T})$$":[158],"O(1/T)":[159],"given":[160],"$$T^{1/4}$$":[169],"T1/4":[170],"$$T^{1/2}$$":[179],"T1/2":[180],"whereTis":[181],"iterations.":[185],"potential":[187],"gains":[188],"can":[190],"be":[191],"achieved":[192],"demonstrated":[196],"empirically":[197],"stochastic":[200],"variational":[201],"inference":[202],"(Latent":[204],"Dirichlet":[205],"Allocation)":[206],"deep":[210],"reinforcement":[211],"learning":[212],"(DRL)":[213],"(advantage":[215],"actor":[216],"critic":[217],"-":[218],"A2C)":[219],"resulting":[220],"algorithms:":[223],"DPSVI":[224],"HSA2C.":[226],"Empirical":[227],"results":[228],"validate":[229],"our":[230],"findings.":[232],"Comparative":[233],"studies":[234],"conducted":[236],"show":[238],"performance":[240],"proposed":[243],"against":[245],"state-of-the-art":[247],"DRL":[248],"algorithms.":[249]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1}],"updated_date":"2026-08-27T14:10:00.468798","created_date":"2025-10-10T00:00:00"}