{"id":"https://openalex.org/W4416707398","doi":"https://doi.org/10.1109/ton.2025.3633713","title":"SIM: Accelerating Distributed DNN Training by Exploring Gradient Similarity","display_name":"SIM: Accelerating Distributed DNN Training by Exploring Gradient Similarity","publication_year":2025,"publication_date":"2025-11-26","ids":{"openalex":"https://openalex.org/W4416707398","doi":"https://doi.org/10.1109/ton.2025.3633713"},"language":null,"primary_location":{"id":"doi:10.1109/ton.2025.3633713","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ton.2025.3633713","pdf_url":null,"source":{"id":"https://openalex.org/S5407042750","display_name":"IEEE Transactions on Networking","issn_l":"2998-4157","issn":["2998-4157"],"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Networking","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/A5101492539","display_name":"Ye Jin","orcid":"https://orcid.org/0000-0001-8087-6333"},"institutions":[{"id":"https://openalex.org/I150807315","display_name":"Guangxi University","ror":"https://ror.org/02c9qn167","country_code":"CN","type":"education","lineage":["https://openalex.org/I150807315"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jin Ye","raw_affiliation_strings":["School of Computer, Electronics and Information, Guangxi University, Nanning, Guangxi, China"],"raw_orcid":"https://orcid.org/0000-0001-8087-6333","affiliations":[{"raw_affiliation_string":"School of Computer, Electronics and Information, Guangxi University, Nanning, Guangxi, China","institution_ids":["https://openalex.org/I150807315"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059077556","display_name":"Yijun Li","orcid":"https://orcid.org/0000-0003-4335-8742"},"institutions":[{"id":"https://openalex.org/I139660479","display_name":"Central South University","ror":"https://ror.org/00f1zfq44","country_code":"CN","type":"education","lineage":["https://openalex.org/I139660479"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yijun Li","raw_affiliation_strings":["School of Computer Science and Engineering, Central South University, Changsha, China"],"raw_orcid":"https://orcid.org/0000-0003-4335-8742","affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Central South University, Changsha, China","institution_ids":["https://openalex.org/I139660479"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100359884","display_name":"Wenliang Li","orcid":"https://orcid.org/0000-0003-2394-2597"},"institutions":[{"id":"https://openalex.org/I150807315","display_name":"Guangxi University","ror":"https://ror.org/02c9qn167","country_code":"CN","type":"education","lineage":["https://openalex.org/I150807315"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenliang Li","raw_affiliation_strings":["School of Computer, Electronics and Information, Guangxi University, Nanning, Guangxi, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer, Electronics and Information, Guangxi University, Nanning, Guangxi, China","institution_ids":["https://openalex.org/I150807315"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101465114","display_name":"Xiaojuan Lu","orcid":"https://orcid.org/0000-0002-4865-8632"},"institutions":[{"id":"https://openalex.org/I139660479","display_name":"Central South University","ror":"https://ror.org/00f1zfq44","country_code":"CN","type":"education","lineage":["https://openalex.org/I139660479"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaojuan Lu","raw_affiliation_strings":["School of Computer Science and Engineering, Central South University, Changsha, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Central South University, Changsha, China","institution_ids":["https://openalex.org/I139660479"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035152634","display_name":"Qichen Su","orcid":"https://orcid.org/0009-0007-7361-343X"},"institutions":[{"id":"https://openalex.org/I139660479","display_name":"Central South University","ror":"https://ror.org/00f1zfq44","country_code":"CN","type":"education","lineage":["https://openalex.org/I139660479"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qichen Su","raw_affiliation_strings":["School of Computer Science and Engineering, Central South University, Changsha, China"],"raw_orcid":"https://orcid.org/0009-0007-7361-343X","affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Central South University, Changsha, China","institution_ids":["https://openalex.org/I139660479"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008483730","display_name":"Jiawei Huang","orcid":"https://orcid.org/0000-0002-7578-4490"},"institutions":[{"id":"https://openalex.org/I139660479","display_name":"Central South University","ror":"https://ror.org/00f1zfq44","country_code":"CN","type":"education","lineage":["https://openalex.org/I139660479"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiawei Huang","raw_affiliation_strings":["School of Computer Science and Engineering, Central South University, Changsha, China"],"raw_orcid":"https://orcid.org/0000-0002-7578-4490","affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Central South University, Changsha, China","institution_ids":["https://openalex.org/I139660479"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100438360","display_name":"Jianxin Wang","orcid":"https://orcid.org/0000-0003-1516-0480"},"institutions":[{"id":"https://openalex.org/I139660479","display_name":"Central South University","ror":"https://ror.org/00f1zfq44","country_code":"CN","type":"education","lineage":["https://openalex.org/I139660479"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jianxin Wang","raw_affiliation_strings":["School of Computer Science and Engineering, Central South University, Changsha, China"],"raw_orcid":"https://orcid.org/0000-0003-1516-0480","affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Central South University, Changsha, China","institution_ids":["https://openalex.org/I139660479"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":{"value":2645,"currency":"USD","value_usd":2645},"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.11982851,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"34","issue":null,"first_page":"1838","last_page":"1850"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.29330000281333923,"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":0.29330000281333923,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.1670999974012375,"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"}},{"id":"https://openalex.org/T10273","display_name":"IoT and Edge/Fog Computing","score":0.04089999943971634,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/convergence","display_name":"Convergence (economics)","score":0.7064999938011169},{"id":"https://openalex.org/keywords/asynchronous-communication","display_name":"Asynchronous communication","score":0.696399986743927},{"id":"https://openalex.org/keywords/testbed","display_name":"Testbed","score":0.6940000057220459},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.6913999915122986},{"id":"https://openalex.org/keywords/synchronization","display_name":"Synchronization (alternating current)","score":0.5927000045776367},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.5782999992370605},{"id":"https://openalex.org/keywords/similarity","display_name":"Similarity (geometry)","score":0.554099977016449},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.4426000118255615}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7307000160217285},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.7064999938011169},{"id":"https://openalex.org/C151319957","wikidata":"https://www.wikidata.org/wiki/Q752739","display_name":"Asynchronous communication","level":2,"score":0.696399986743927},{"id":"https://openalex.org/C31395832","wikidata":"https://www.wikidata.org/wiki/Q1318674","display_name":"Testbed","level":2,"score":0.6940000057220459},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.6913999915122986},{"id":"https://openalex.org/C2778562939","wikidata":"https://www.wikidata.org/wiki/Q1298791","display_name":"Synchronization (alternating current)","level":3,"score":0.5927000045776367},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.5782999992370605},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.554099977016449},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47130000591278076},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.4426000118255615},{"id":"https://openalex.org/C2779582901","wikidata":"https://www.wikidata.org/wiki/Q21013010","display_name":"Distributed learning","level":2,"score":0.3917999863624573},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.3790000081062317},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.36469998955726624},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.36250001192092896},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.3416000008583069},{"id":"https://openalex.org/C115680565","wikidata":"https://www.wikidata.org/wiki/Q5977448","display_name":"Gradient method","level":2,"score":0.32760000228881836},{"id":"https://openalex.org/C2777131613","wikidata":"https://www.wikidata.org/wiki/Q18394147","display_name":"Retard","level":2,"score":0.3203999996185303},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3197000026702881},{"id":"https://openalex.org/C130120984","wikidata":"https://www.wikidata.org/wiki/Q2835898","display_name":"Distributed algorithm","level":2,"score":0.298799991607666},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.29490000009536743},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.27649998664855957},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.266400009393692},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2531000077724457}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ton.2025.3633713","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ton.2025.3633713","pdf_url":null,"source":{"id":"https://openalex.org/S5407042750","display_name":"IEEE Transactions on Networking","issn_l":"2998-4157","issn":["2998-4157"],"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Networking","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1946796335","display_name":null,"funder_award_id":"U24A20245","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G2099920728","display_name":null,"funder_award_id":"2024RC1005","funder_id":"https://openalex.org/F4320330206","funder_display_name":"Science and Technology Program of Hunan Province"},{"id":"https://openalex.org/G3358112784","display_name":null,"funder_award_id":"62132022","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G921486363","display_name":null,"funder_award_id":"U23A20313","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/F4320330206","display_name":"Science and Technology Program of Hunan Province","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":19,"referenced_works":["https://openalex.org/W2045271686","https://openalex.org/W2083842231","https://openalex.org/W2194775991","https://openalex.org/W2580688187","https://openalex.org/W2618530766","https://openalex.org/W2920397365","https://openalex.org/W2963703197","https://openalex.org/W2965481041","https://openalex.org/W2975712713","https://openalex.org/W3016249977","https://openalex.org/W3047357290","https://openalex.org/W3086105743","https://openalex.org/W3109982791","https://openalex.org/W3134742951","https://openalex.org/W3155611867","https://openalex.org/W3166384310","https://openalex.org/W4283219828","https://openalex.org/W4287332481","https://openalex.org/W4313417875"],"related_works":[],"abstract_inverted_index":{"Synchronous":[0],"stochastic":[1],"gradient":[2,102],"descent":[3],"(SSGD)":[4],"has":[5],"been":[6],"widely":[7],"used":[8],"in":[9],"distributed":[10],"deep":[11,148],"learning.":[12],"However,":[13],"since":[14],"the":[15,82,94,101,111,115,142],"local":[16,107],"gradients":[17],"need":[18],"to":[19,54,153],"be":[20],"shared":[21],"among":[22],"workers":[23,98],"at":[24],"every":[25],"iteration,":[26],"SSGD":[27],"performance":[28],"is":[29],"significantly":[30],"influenced":[31],"by":[32,36,151],"network":[33,85],"bottlenecks":[34,86],"caused":[35],"either":[37],"heterogeneous":[38],"environment":[39],"or":[40],"bandwidth":[41],"contention.":[42],"To":[43],"solve":[44],"this":[45,69],"problem,":[46],"asynchronous":[47],"parallel":[48],"(ASP)":[49],"strategy":[50],"allows":[51],"each":[52],"worker":[53],"update":[55],"parameters":[56],"independently":[57],"without":[58],"synchronization,":[59],"while":[60],"suffering":[61],"from":[62],"accuracy":[63],"loss":[64],"and":[65,87,106,125,132],"convergence":[66,89,123,143],"inefficiency.":[67],"In":[68],"paper,":[70],"we":[71],"propose":[72],"a":[73,119],"novel":[74],"similarity-based":[75],"synchronization":[76],"scheme":[77],"called":[78],"SIM,":[79],"which":[80],"mitigates":[81],"impact":[83],"of":[84,96,122,145],"ensures":[88],"efficiency.":[90],"Specifically,":[91],"SIM":[92,140],"reduces":[93,141],"number":[95],"aggregation":[97],"based":[99],"on":[100,130],"similarity":[103],"between":[104],"global":[105],"gradients,":[108],"therefore":[109],"shrinking":[110],"waiting":[112],"time":[113,144],"for":[114],"stragglers.":[116],"We":[117],"provide":[118],"theoretical":[120],"analysis":[121],"efficiency":[124],"conduct":[126],"large-scale":[127],"testbed":[128],"experiments":[129],"CIFAR-10":[131],"SQUAD":[133],"dataset.":[134],"The":[135],"experimental":[136],"results":[137],"show":[138],"that":[139],"four":[146],"classical":[147],"learning":[149],"models":[150],"up":[152],"40%.":[154]},"counts_by_year":[],"updated_date":"2026-08-21T09:56:20.448147","created_date":"2025-11-27T00:00:00"}