{"id":"https://openalex.org/W3155611867","doi":"https://doi.org/10.1109/infocom42981.2021.9488815","title":"Live Gradient Compensation for Evading Stragglers in Distributed Learning","display_name":"Live Gradient Compensation for Evading Stragglers in Distributed Learning","publication_year":2021,"publication_date":"2021-05-10","ids":{"openalex":"https://openalex.org/W3155611867","doi":"https://doi.org/10.1109/infocom42981.2021.9488815","mag":"3155611867"},"language":"en","primary_location":{"id":"doi:10.1109/infocom42981.2021.9488815","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom42981.2021.9488815","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE INFOCOM 2021 - IEEE Conference on Computer Communications","raw_type":"proceedings-article"},"type":"conference-paper","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/A5101613464","display_name":"Jian Xu","orcid":"https://orcid.org/0000-0001-6201-9215"},"institutions":[{"id":"https://openalex.org/I4210114105","display_name":"Tsinghua\u2013Berkeley Shenzhen Institute","ror":"https://ror.org/02hhwwz98","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210114105","https://openalex.org/I95457486","https://openalex.org/I99065089"]},{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jian Xu","raw_affiliation_strings":["Tsinghua-Berkeley Shenzhen Institute, Tsinghua University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua-Berkeley Shenzhen Institute, Tsinghua University","institution_ids":["https://openalex.org/I4210114105","https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088293566","display_name":"Shao\u2010Lun Huang","orcid":"https://orcid.org/0000-0003-2827-4022"},"institutions":[{"id":"https://openalex.org/I4210114105","display_name":"Tsinghua\u2013Berkeley Shenzhen Institute","ror":"https://ror.org/02hhwwz98","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210114105","https://openalex.org/I95457486","https://openalex.org/I99065089"]},{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shao-Lun Huang","raw_affiliation_strings":["Tsinghua-Berkeley Shenzhen Institute, Tsinghua University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua-Berkeley Shenzhen Institute, Tsinghua University","institution_ids":["https://openalex.org/I4210114105","https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035185924","display_name":"Linqi Song","orcid":"https://orcid.org/0000-0003-2756-4984"},"institutions":[{"id":"https://openalex.org/I168719708","display_name":"City University of Hong Kong","ror":"https://ror.org/03q8dnn23","country_code":"HK","type":"education","lineage":["https://openalex.org/I168719708"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Linqi Song","raw_affiliation_strings":["City University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"City University of Hong Kong","institution_ids":["https://openalex.org/I168719708"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5018464968","display_name":"Tian Lan","orcid":"https://orcid.org/0000-0003-3010-8090"},"institutions":[{"id":"https://openalex.org/I193531525","display_name":"George Washington University","ror":"https://ror.org/00y4zzh67","country_code":"US","type":"education","lineage":["https://openalex.org/I193531525"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tian Lan","raw_affiliation_strings":["George Washington University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"George Washington University","institution_ids":["https://openalex.org/I193531525"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":27,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"10"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9995999932289124,"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":0.9995999932289124,"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":0.9988999962806702,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9980000257492065,"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/computer-science","display_name":"Computer science","score":0.7953333854675293},{"id":"https://openalex.org/keywords/compensation","display_name":"Compensation (psychology)","score":0.6654958724975586},{"id":"https://openalex.org/keywords/convergence","display_name":"Convergence (economics)","score":0.5731875896453857},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.4197048544883728},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3717186450958252},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.364352285861969}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7953333854675293},{"id":"https://openalex.org/C2780023022","wikidata":"https://www.wikidata.org/wiki/Q1338171","display_name":"Compensation (psychology)","level":2,"score":0.6654958724975586},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.5731875896453857},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.4197048544883728},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3717186450958252},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.364352285861969},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C50522688","wikidata":"https://www.wikidata.org/wiki/Q189833","display_name":"Economic growth","level":1,"score":0.0},{"id":"https://openalex.org/C11171543","wikidata":"https://www.wikidata.org/wiki/Q41630","display_name":"Psychoanalysis","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/infocom42981.2021.9488815","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom42981.2021.9488815","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE INFOCOM 2021 - IEEE Conference on Computer Communications","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.6800000071525574,"display_name":"Decent work and economic growth","id":"https://metadata.un.org/sdg/8"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":76,"referenced_works":["https://openalex.org/W778657980","https://openalex.org/W1686810756","https://openalex.org/W1982063824","https://openalex.org/W2087402357","https://openalex.org/W2112796928","https://openalex.org/W2127941149","https://openalex.org/W2138243089","https://openalex.org/W2160815625","https://openalex.org/W2163605009","https://openalex.org/W2168231600","https://openalex.org/W2184045248","https://openalex.org/W2268702383","https://openalex.org/W2336650964","https://openalex.org/W2622263826","https://openalex.org/W2741736887","https://openalex.org/W2751733006","https://openalex.org/W2787128677","https://openalex.org/W2792603880","https://openalex.org/W2804095446","https://openalex.org/W2807006176","https://openalex.org/W2890924858","https://openalex.org/W2906936451","https://openalex.org/W2914855042","https://openalex.org/W2944542720","https://openalex.org/W2945336802","https://openalex.org/W2946815927","https://openalex.org/W2949585412","https://openalex.org/W2949934631","https://openalex.org/W2955213239","https://openalex.org/W2962873681","https://openalex.org/W2963120442","https://openalex.org/W2963361660","https://openalex.org/W2963433607","https://openalex.org/W2963903325","https://openalex.org/W2964120628","https://openalex.org/W2964160151","https://openalex.org/W2964352239","https://openalex.org/W2971342441","https://openalex.org/W2976713597","https://openalex.org/W2989289980","https://openalex.org/W2995653155","https://openalex.org/W3001445294","https://openalex.org/W3016249977","https://openalex.org/W3023838981","https://openalex.org/W3037047862","https://openalex.org/W3100515187","https://openalex.org/W3117733161","https://openalex.org/W3118608800","https://openalex.org/W4287759323","https://openalex.org/W6622473587","https://openalex.org/W6637373629","https://openalex.org/W6679154944","https://openalex.org/W6684191040","https://openalex.org/W6684859321","https://openalex.org/W6703420464","https://openalex.org/W6739622702","https://openalex.org/W6742061381","https://openalex.org/W6743655711","https://openalex.org/W6748342376","https://openalex.org/W6749257597","https://openalex.org/W6749835716","https://openalex.org/W6751815326","https://openalex.org/W6752029299","https://openalex.org/W6754416507","https://openalex.org/W6755251367","https://openalex.org/W6757392376","https://openalex.org/W6759084776","https://openalex.org/W6760157594","https://openalex.org/W6762421227","https://openalex.org/W6762930437","https://openalex.org/W6763393679","https://openalex.org/W6764309476","https://openalex.org/W6765541894","https://openalex.org/W6772835049","https://openalex.org/W6787972765","https://openalex.org/W6788047648"],"related_works":["https://openalex.org/W2051487156","https://openalex.org/W2073681303","https://openalex.org/W2379084545","https://openalex.org/W2053286651","https://openalex.org/W2181743346","https://openalex.org/W2187401768","https://openalex.org/W2181413294","https://openalex.org/W2989452537","https://openalex.org/W2052122378","https://openalex.org/W2544423928"],"abstract_inverted_index":{"The":[0,126],"training":[1,110,139],"efficiency":[2],"of":[3,101,153,168],"distributed":[4,22],"learning":[5,23,64],"systems":[6],"is":[7,19],"vulnerable":[8],"to":[9,53,62,112,132],"stragglers,":[10,33,60],"namely,":[11],"those":[12],"slow":[13],"worker":[14],"nodes.":[15],"A":[16],"naive":[17],"strategy":[18,52],"performing":[20],"the":[21,26,55,68,86,118,145,162,166],"by":[24,143],"incor-porating":[25],"fastest":[27],"K":[28],"workers":[29],"and":[30,66,81,141,161],"ignoring":[31],"these":[32],"which":[34,84],"may":[35],"induce":[36],"high":[37],"deviation":[38],"for":[39,104],"non-IID":[40,109],"data.":[41],"To":[42],"tackle":[43],"this,":[44],"we":[45,96],"develop":[46],"a":[47,134],"Live":[48],"Gradient":[49],"Compensation":[50],"(LGC)":[51],"incorporate":[54],"one-step":[56],"delayed":[57],"gradients":[58],"from":[59],"aiming":[61],"accelerate":[63],"process":[65],"utilize":[67],"stragglers":[69],"simultaneously.":[70],"In":[71,94],"LGC":[72],"framework,":[73],"mini-batch":[74],"data":[75,111],"are":[76,159],"divided":[77],"into":[78],"smaller":[79],"blocks":[80],"processed":[82],"separately,":[83],"makes":[85],"gradient":[87],"computed":[88],"based":[89],"on":[90,156],"partial":[91],"work":[92],"accessible.":[93],"addition,":[95],"provide":[97],"theoretical":[98,127],"convergence":[99,120],"analysis":[100],"our":[102,169],"algorithm":[103],"non-convex":[105],"optimization":[106],"problem":[107],"under":[108],"show":[113],"that":[114],"LGC-SGD":[115],"has":[116],"almost":[117],"same":[119],"error":[121,142],"as":[122],"full":[123],"synchronous":[124],"SGD.":[125],"results":[128,164],"also":[129],"allow":[130],"us":[131],"quantify":[133],"novel":[135],"tradeoff":[136],"in":[137],"minimizing":[138],"time":[140],"selecting":[144],"optimal":[146],"straggler":[147],"threshold.":[148],"Finally,":[149],"extensive":[150],"simulation":[151],"experiments":[152],"image":[154],"classification":[155],"CIFAR-10":[157],"dataset":[158],"conducted,":[160],"numerical":[163],"demonstrate":[165],"effectiveness":[167],"proposed":[170],"strategy.":[171]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":10},{"year":2023,"cited_by_count":3},{"year":2022,"cited_by_count":5},{"year":2021,"cited_by_count":4}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}