{"id":"https://openalex.org/W2991820076","doi":"https://doi.org/10.1109/allerton.2019.8919791","title":"Distributed SGD Generalizes Well Under Asynchrony","display_name":"Distributed SGD Generalizes Well Under Asynchrony","publication_year":2019,"publication_date":"2019-09-01","ids":{"openalex":"https://openalex.org/W2991820076","doi":"https://doi.org/10.1109/allerton.2019.8919791","mag":"2991820076"},"language":"en","primary_location":{"id":"doi:10.1109/allerton.2019.8919791","is_oa":false,"landing_page_url":"https://doi.org/10.1109/allerton.2019.8919791","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2019 57th Annual Allerton Conference on Communication, Control, and Computing (Allerton)","raw_type":"proceedings-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/A5046845999","display_name":"Jayanth Reddy Regatti","orcid":"https://orcid.org/0000-0001-8150-2288"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Jayanth Regatti","raw_affiliation_strings":["Ohio State University,ECE department,Columbus,OH","ECE department, Ohio State University, Columbus, OH"],"affiliations":[{"raw_affiliation_string":"Ohio State University,ECE department,Columbus,OH","institution_ids":["https://openalex.org/I52357470"]},{"raw_affiliation_string":"ECE department, Ohio State University, Columbus, OH","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077624035","display_name":"Gaurav Tendolkar","orcid":null},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gaurav Tendolkar","raw_affiliation_strings":["Ohio State University,ECE department,Columbus,OH","ECE department, Ohio State University, Columbus, OH"],"affiliations":[{"raw_affiliation_string":"Ohio State University,ECE department,Columbus,OH","institution_ids":["https://openalex.org/I52357470"]},{"raw_affiliation_string":"ECE department, Ohio State University, Columbus, OH","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101704070","display_name":"Yi Zhou","orcid":"https://orcid.org/0000-0002-3982-9145"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yi Zhou","raw_affiliation_strings":["Ohio State University,ECE department,Columbus,OH","ECE department, Ohio State University, Columbus, OH"],"affiliations":[{"raw_affiliation_string":"Ohio State University,ECE department,Columbus,OH","institution_ids":["https://openalex.org/I52357470"]},{"raw_affiliation_string":"ECE department, Ohio State University, Columbus, OH","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017906439","display_name":"Abhishek Gupta","orcid":"https://orcid.org/0000-0003-1117-325X"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Abhishek Gupta","raw_affiliation_strings":["Ohio State University,ECE department,Columbus,OH","ECE department, Ohio State University, Columbus, OH"],"affiliations":[{"raw_affiliation_string":"Ohio State University,ECE department,Columbus,OH","institution_ids":["https://openalex.org/I52357470"]},{"raw_affiliation_string":"ECE department, Ohio State University, Columbus, OH","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100384384","display_name":"Yingbin Liang","orcid":"https://orcid.org/0000-0003-2631-4262"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yingbin Liang","raw_affiliation_strings":["Ohio State University,ECE department,Columbus,OH","ECE department, Ohio State University, Columbus, OH"],"affiliations":[{"raw_affiliation_string":"Ohio State University,ECE department,Columbus,OH","institution_ids":["https://openalex.org/I52357470"]},{"raw_affiliation_string":"ECE department, Ohio State University, Columbus, OH","institution_ids":["https://openalex.org/I52357470"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5046845999"],"corresponding_institution_ids":["https://openalex.org/I52357470"],"apc_list":null,"apc_paid":null,"fwci":0.1443,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.60258572,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"863","last_page":"870"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9998000264167786,"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":0.9998000264167786,"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9937999844551086,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.9912999868392944,"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/asynchrony","display_name":"Asynchrony (computer programming)","score":0.8606441020965576},{"id":"https://openalex.org/keywords/asynchronous-communication","display_name":"Asynchronous communication","score":0.8435158133506775},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8013226985931396},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.733502209186554},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.7275323271751404},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.6685500144958496},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.6343185901641846},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.6032472848892212},{"id":"https://openalex.org/keywords/distributed-algorithm","display_name":"Distributed algorithm","score":0.5673706531524658},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.5667977333068848},{"id":"https://openalex.org/keywords/independent-and-identically-distributed-random-variables","display_name":"Independent and identically distributed random variables","score":0.4947342574596405},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.24513936042785645},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.2049233317375183},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.19284409284591675},{"id":"https://openalex.org/keywords/random-variable","display_name":"Random variable","score":0.11653971672058105},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.10426318645477295},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.09500139951705933}],"concepts":[{"id":"https://openalex.org/C2779019669","wikidata":"https://www.wikidata.org/wiki/Q25203946","display_name":"Asynchrony (computer programming)","level":3,"score":0.8606441020965576},{"id":"https://openalex.org/C151319957","wikidata":"https://www.wikidata.org/wiki/Q752739","display_name":"Asynchronous communication","level":2,"score":0.8435158133506775},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8013226985931396},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.733502209186554},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.7275323271751404},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6685500144958496},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.6343185901641846},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.6032472848892212},{"id":"https://openalex.org/C130120984","wikidata":"https://www.wikidata.org/wiki/Q2835898","display_name":"Distributed algorithm","level":2,"score":0.5673706531524658},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.5667977333068848},{"id":"https://openalex.org/C141513077","wikidata":"https://www.wikidata.org/wiki/Q378542","display_name":"Independent and identically distributed random variables","level":3,"score":0.4947342574596405},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.24513936042785645},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2049233317375183},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.19284409284591675},{"id":"https://openalex.org/C122123141","wikidata":"https://www.wikidata.org/wiki/Q176623","display_name":"Random variable","level":2,"score":0.11653971672058105},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.10426318645477295},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.09500139951705933},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.0},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/allerton.2019.8919791","is_oa":false,"landing_page_url":"https://doi.org/10.1109/allerton.2019.8919791","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2019 57th Annual Allerton Conference on Communication, Control, and Computing (Allerton)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":53,"referenced_works":["https://openalex.org/W27434444","https://openalex.org/W1603765807","https://openalex.org/W1811750039","https://openalex.org/W1978564754","https://openalex.org/W2016937404","https://openalex.org/W2049469158","https://openalex.org/W2060393849","https://openalex.org/W2061689074","https://openalex.org/W2083842231","https://openalex.org/W2096544401","https://openalex.org/W2100007248","https://openalex.org/W2131542408","https://openalex.org/W2132737349","https://openalex.org/W2136351519","https://openalex.org/W2138243089","https://openalex.org/W2154834860","https://openalex.org/W2166706236","https://openalex.org/W2168231600","https://openalex.org/W2351849545","https://openalex.org/W2569020760","https://openalex.org/W2604451472","https://openalex.org/W2606474801","https://openalex.org/W2616811598","https://openalex.org/W2736970300","https://openalex.org/W2788714855","https://openalex.org/W2949950906","https://openalex.org/W2962950660","https://openalex.org/W2963094221","https://openalex.org/W2963306214","https://openalex.org/W2963361660","https://openalex.org/W2963794891","https://openalex.org/W3005141333","https://openalex.org/W4285719527","https://openalex.org/W4295116634","https://openalex.org/W6638214083","https://openalex.org/W6662929504","https://openalex.org/W6675069819","https://openalex.org/W6679393576","https://openalex.org/W6679563152","https://openalex.org/W6680127587","https://openalex.org/W6680402377","https://openalex.org/W6684249991","https://openalex.org/W6684859321","https://openalex.org/W6692488971","https://openalex.org/W6705795681","https://openalex.org/W6735717691","https://openalex.org/W6736526899","https://openalex.org/W6741472820","https://openalex.org/W6745375793","https://openalex.org/W6748282221","https://openalex.org/W6748287519","https://openalex.org/W6749257597","https://openalex.org/W7034108470"],"related_works":["https://openalex.org/W2595172197","https://openalex.org/W2534656837","https://openalex.org/W2084856301","https://openalex.org/W2127970246","https://openalex.org/W2885125400","https://openalex.org/W1989889224","https://openalex.org/W4382618745","https://openalex.org/W2974995819","https://openalex.org/W4288026461","https://openalex.org/W1489256244"],"abstract_inverted_index":{"The":[0,50],"performance":[1,39],"of":[2,40,53,135],"fully":[3],"synchronized":[4],"distributed":[5,20,47,95,125,137],"systems":[6,21],"has":[7],"faced":[8],"a":[9,24,46,69],"bottleneck":[10],"due":[11,27],"to":[12,28,65,73,113],"the":[13,37,75,78,81,88,105,115,124,133,136,141],"big":[14],"data":[15,102],"trend,":[16],"under":[17],"which":[18,61],"asynchronous":[19,48,96],"are":[22,62],"becoming":[23],"major":[25],"popularity":[26],"their":[29],"powerful":[30],"scalability.":[31],"In":[32,108],"this":[33],"paper,":[34],"we":[35,92,119,146],"study":[36],"generalization":[38,143],"stochastic":[41,59],"gradient":[42],"descent":[43],"(SGD)":[44],"on":[45,68],"system.":[49,126],"system":[51,82],"consists":[52],"multiple":[54],"worker":[55],"machines":[56],"that":[57,94],"compute":[58],"gradients":[60],"further":[63],"sent":[64],"and":[66,77,139],"aggregated":[67],"common":[70],"parameter":[71],"server":[72],"update":[74],"variables,":[76],"communication":[79],"in":[80,104,123],"suffers":[83],"from":[84],"possible":[85],"delays.":[86],"Under":[87],"algorithm":[89,138],"stability":[90,134],"framework,":[91],"prove":[93],"SGD":[97],"generalizes":[98],"well":[99],"given":[100],"enough":[101],"samples":[103],"training":[106],"optimization.":[107],"particular,":[109],"our":[110,148],"results":[111],"suggest":[112],"reduce":[114],"learning":[116,129],"rate":[117,130],"as":[118],"allow":[120],"more":[121],"asynchrony":[122],"Such":[127],"adaptive":[128],"strategy":[131],"improves":[132],"reduces":[140],"corresponding":[142],"error.":[144],"Then,":[145],"confirm":[147],"theoretical":[149],"findings":[150],"via":[151],"numerical":[152],"experiments.":[153]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2021,"cited_by_count":1}],"updated_date":"2026-04-02T13:48:15.688549","created_date":"2025-10-10T00:00:00"}
