{"id":"https://openalex.org/W3110023104","doi":"https://doi.org/10.1587/transinf.2020pap0007","title":"An Efficient Method for Training Deep Learning Networks Distributed","display_name":"An Efficient Method for Training Deep Learning Networks Distributed","publication_year":2020,"publication_date":"2020-11-30","ids":{"openalex":"https://openalex.org/W3110023104","doi":"https://doi.org/10.1587/transinf.2020pap0007","mag":"3110023104"},"language":"en","primary_location":{"id":"doi:10.1587/transinf.2020pap0007","is_oa":true,"landing_page_url":"https://doi.org/10.1587/transinf.2020pap0007","pdf_url":"https://www.jstage.jst.go.jp/article/transinf/E103.D/12/E103.D_2020PAP0007/_pdf","source":{"id":"https://openalex.org/S2486202937","display_name":"IEICE Transactions on Information and Systems","issn_l":"0916-8532","issn":["0916-8532","1745-1361"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4320800604","host_organization_name":"Institute of Electronics, Information and Communication Engineers","host_organization_lineage":["https://openalex.org/P4320800604"],"host_organization_lineage_names":["Institute of Electronics, Information and Communication Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEICE Transactions on Information and Systems","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://www.jstage.jst.go.jp/article/transinf/E103.D/12/E103.D_2020PAP0007/_pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100455049","display_name":"Chenxu Wang","orcid":"https://orcid.org/0000-0003-1276-5759"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chenxu WANG","raw_affiliation_strings":["School of Computer Science, National University of Defense Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science, National University of Defense Technology","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101633465","display_name":"Yutong Lu","orcid":"https://orcid.org/0000-0001-5315-3375"},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]},{"id":"https://openalex.org/I4210112812","display_name":"National Supercomputing Center in Shenzhen","ror":"https://ror.org/02291hh73","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210112812"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yutong LU","raw_affiliation_strings":["National Supercomputer Center in Guangzhou","School of Data and Computer Science, Sun Yat-sen University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National Supercomputer Center in Guangzhou","institution_ids":["https://openalex.org/I4210112812"]},{"raw_affiliation_string":"School of Data and Computer Science, Sun Yat-sen University","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101483479","display_name":"Zhiguang Chen","orcid":"https://orcid.org/0000-0002-9318-5715"},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]},{"id":"https://openalex.org/I4210112812","display_name":"National Supercomputing Center in Shenzhen","ror":"https://ror.org/02291hh73","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210112812"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhiguang CHEN","raw_affiliation_strings":["National Supercomputer Center in Guangzhou","School of Data and Computer Science, Sun Yat-sen University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National Supercomputer Center in Guangzhou","institution_ids":["https://openalex.org/I4210112812"]},{"raw_affiliation_string":"School of Data and Computer Science, Sun Yat-sen University","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100608756","display_name":"Junnan Li","orcid":"https://orcid.org/0000-0002-1405-2034"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Junnan LI","raw_affiliation_strings":["School of Computer Science, National University of Defense Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science, National University of Defense Technology","institution_ids":["https://openalex.org/I170215575"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.1917,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.50867673,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":94},"biblio":{"volume":"E103.D","issue":"12","first_page":"2444","last_page":"2456"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9987999796867371,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9987999796867371,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","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"}},{"id":"https://openalex.org/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9975000023841858,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.9240300059318542},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.7208611369132996},{"id":"https://openalex.org/keywords/synchronization","display_name":"Synchronization (alternating current)","score":0.6535696983337402},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.6431340575218201},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.5802531242370605},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5598442554473877},{"id":"https://openalex.org/keywords/scheme","display_name":"Scheme (mathematics)","score":0.5289509296417236},{"id":"https://openalex.org/keywords/interconnection","display_name":"Interconnection","score":0.4520138204097748},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.44687703251838684},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.3792111277580261},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3710576295852661},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.32408249378204346},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.22191861271858215}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.9240300059318542},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.7208611369132996},{"id":"https://openalex.org/C2778562939","wikidata":"https://www.wikidata.org/wiki/Q1298791","display_name":"Synchronization (alternating current)","level":3,"score":0.6535696983337402},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.6431340575218201},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.5802531242370605},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5598442554473877},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.5289509296417236},{"id":"https://openalex.org/C123745756","wikidata":"https://www.wikidata.org/wiki/Q1665949","display_name":"Interconnection","level":2,"score":0.4520138204097748},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.44687703251838684},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3792111277580261},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3710576295852661},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.32408249378204346},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.22191861271858215},{"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/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1587/transinf.2020pap0007","is_oa":true,"landing_page_url":"https://doi.org/10.1587/transinf.2020pap0007","pdf_url":"https://www.jstage.jst.go.jp/article/transinf/E103.D/12/E103.D_2020PAP0007/_pdf","source":{"id":"https://openalex.org/S2486202937","display_name":"IEICE Transactions on Information and Systems","issn_l":"0916-8532","issn":["0916-8532","1745-1361"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4320800604","host_organization_name":"Institute of Electronics, Information and Communication Engineers","host_organization_lineage":["https://openalex.org/P4320800604"],"host_organization_lineage_names":["Institute of Electronics, Information and Communication Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEICE Transactions on Information and Systems","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1587/transinf.2020pap0007","is_oa":true,"landing_page_url":"https://doi.org/10.1587/transinf.2020pap0007","pdf_url":"https://www.jstage.jst.go.jp/article/transinf/E103.D/12/E103.D_2020PAP0007/_pdf","source":{"id":"https://openalex.org/S2486202937","display_name":"IEICE Transactions on Information and Systems","issn_l":"0916-8532","issn":["0916-8532","1745-1361"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4320800604","host_organization_name":"Institute of Electronics, Information and Communication Engineers","host_organization_lineage":["https://openalex.org/P4320800604"],"host_organization_lineage_names":["Institute of Electronics, Information and Communication Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEICE Transactions on Information and Systems","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3110023104.pdf","grobid_xml":"https://content.openalex.org/works/W3110023104.grobid-xml"},"referenced_works_count":35,"referenced_works":["https://openalex.org/W1442374986","https://openalex.org/W1598866093","https://openalex.org/W1641327695","https://openalex.org/W1686810756","https://openalex.org/W2081612620","https://openalex.org/W2097117768","https://openalex.org/W2112796928","https://openalex.org/W2117539524","https://openalex.org/W2163605009","https://openalex.org/W2168231600","https://openalex.org/W2170135819","https://openalex.org/W2194775991","https://openalex.org/W2287011250","https://openalex.org/W2336650964","https://openalex.org/W2407022425","https://openalex.org/W2512138407","https://openalex.org/W2553581924","https://openalex.org/W2579247884","https://openalex.org/W2580688187","https://openalex.org/W2622263826","https://openalex.org/W2728256789","https://openalex.org/W2740067745","https://openalex.org/W2745269232","https://openalex.org/W2747482047","https://openalex.org/W2749988060","https://openalex.org/W2759465730","https://openalex.org/W2774000609","https://openalex.org/W2775776326","https://openalex.org/W2787405437","https://openalex.org/W2790601347","https://openalex.org/W2899133614","https://openalex.org/W2912759934","https://openalex.org/W2951894856","https://openalex.org/W2963042536","https://openalex.org/W3118608800"],"related_works":["https://openalex.org/W2350861899","https://openalex.org/W2019484008","https://openalex.org/W4285351355","https://openalex.org/W1521353230","https://openalex.org/W2391167130","https://openalex.org/W1522862521","https://openalex.org/W2092071486","https://openalex.org/W4283067488","https://openalex.org/W2131630752","https://openalex.org/W2573708829"],"abstract_inverted_index":{"Training":[0],"deep":[1],"learning":[2],"(DL)":[3],"is":[4],"a":[5,10,34,59,72,96],"computationally":[6],"intensive":[7],"process;":[8],"as":[9,129,131],"result,":[11],"training":[12],"time":[13],"can":[14,47,81,102,179],"become":[15],"so":[16],"long":[17],"that":[18,156],"it":[19],"impedes":[20],"the":[21,110,120,135,146,169],"development":[22],"of":[23,37,85,109,138],"DL.":[24],"High":[25],"performance":[26,161],"computing":[27,38,181],"clusters,":[28],"especially":[29],"supercomputers,":[30],"are":[31],"equipped":[32],"with":[33,66,168],"large":[35],"amount":[36],"resources,":[39,41],"storage":[40],"and":[42,52,88],"efficient":[43],"interconnection":[44],"ability,":[45],"which":[46,80,101],"train":[48,62],"DL":[49,63],"networks":[50,64],"better":[51],"faster.":[53],"In":[54],"this":[55],"paper,":[56],"we":[57,70,94,118,144],"propose":[58,71],"method":[60],"to":[61,133,164],"distributed":[65,171],"high":[67,136],"efficiency.":[68,92],"First,":[69],"hierarchical":[73,174],"synchronous":[74,175],"Stochastic":[75],"Gradient":[76],"Descent":[77],"(SGD)":[78],"strategy,":[79,172],"make":[82],"full":[83],"use":[84],"hardware":[86],"resources":[87],"greatly":[89],"increase":[90,180],"computational":[91],"Second,":[93],"present":[95],"two-level":[97],"parameter":[98],"synchronization":[99],"scheme":[100],"reduce":[103],"communication":[104],"overhead":[105,137],"by":[106,123,183],"transmitting":[107],"parameters":[108],"first":[111],"layer":[112],"models":[113],"in":[114],"shared":[115],"memory.":[116],"Third,":[117],"optimize":[119],"parallel":[121],"I/O":[122],"making":[124],"each":[125],"reader":[126],"read":[127],"data":[128,140],"continuously":[130],"possible":[132],"avoid":[134],"discontinuous":[139],"reading.":[141],"At":[142],"last,":[143],"integrate":[145],"LARS":[147],"algorithm":[148],"into":[149],"our":[150,157,173],"system.":[151],"The":[152],"experimental":[153],"results":[154],"demonstrate":[155],"approach":[158],"has":[159],"tremendous":[160],"advantages":[162],"relative":[163],"unoptimized":[165],"methods.":[166],"Compared":[167],"native":[170],"SGD":[176],"strategy":[177],"(HSGD)":[178],"efficiency":[182],"about":[184],"20":[185],"times.":[186]},"counts_by_year":[{"year":2023,"cited_by_count":1},{"year":2021,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}