{"id":"https://openalex.org/W2963768958","doi":"https://doi.org/10.1145/3339186.3339203","title":"Reducing global reductions in large-scale distributed training","display_name":"Reducing global reductions in large-scale distributed training","publication_year":2019,"publication_date":"2019-07-22","ids":{"openalex":"https://openalex.org/W2963768958","doi":"https://doi.org/10.1145/3339186.3339203","mag":"2963768958"},"language":"en","primary_location":{"id":"doi:10.1145/3339186.3339203","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3339186.3339203","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Workshop Proceedings of the 48th International Conference on Parallel Processing","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/A5089801879","display_name":"Guojing Cong","orcid":"https://orcid.org/0000-0003-0850-7714"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Guojing Cong","raw_affiliation_strings":["IBM TJ Watson Research Center, Yorktown Heights, USA"],"affiliations":[{"raw_affiliation_string":"IBM TJ Watson Research Center, Yorktown Heights, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101738461","display_name":"Chih-Chieh Yang","orcid":"https://orcid.org/0000-0003-2816-4464"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chih-Chieh Yang","raw_affiliation_strings":["IBM TJ Watson Research Center, Yorktown Heights, USA"],"affiliations":[{"raw_affiliation_string":"IBM TJ Watson Research Center, Yorktown Heights, USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5077610457","display_name":"Fan Zhou","orcid":"https://orcid.org/0000-0003-1736-2641"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fan Zhou","raw_affiliation_strings":["Baidu USA, Sunnyville, USA"],"affiliations":[{"raw_affiliation_string":"Baidu USA, Sunnyville, USA","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5089801879"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.28,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.65022249,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":94},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"9"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9998999834060669,"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":0.9998999834060669,"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":0.9990000128746033,"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/T12676","display_name":"Machine Learning and ELM","score":0.9984999895095825,"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":0.7416800260543823},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.7371985912322998},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.6656829118728638},{"id":"https://openalex.org/keywords/convergence","display_name":"Convergence (economics)","score":0.616097092628479},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.6120578646659851},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.5700010657310486},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.4871944785118103},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.48362627625465393},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.47117286920547485},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.4349609315395355},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.40366899967193604},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.358887642621994},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3494434356689453},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1740015745162964}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7416800260543823},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.7371985912322998},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.6656829118728638},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.616097092628479},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.6120578646659851},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.5700010657310486},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.4871944785118103},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.48362627625465393},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.47117286920547485},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.4349609315395355},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.40366899967193604},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.358887642621994},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3494434356689453},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1740015745162964},{"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/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"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/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","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},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3339186.3339203","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3339186.3339203","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Workshop Proceedings of the 48th International Conference on Parallel Processing","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":35,"referenced_works":["https://openalex.org/W778657980","https://openalex.org/W1516903196","https://openalex.org/W1686810756","https://openalex.org/W1972711404","https://openalex.org/W1992208280","https://openalex.org/W1994616650","https://openalex.org/W2130062883","https://openalex.org/W2132211083","https://openalex.org/W2146502635","https://openalex.org/W2162746441","https://openalex.org/W2166706236","https://openalex.org/W2168231600","https://openalex.org/W2497446532","https://openalex.org/W2525778437","https://openalex.org/W2566079294","https://openalex.org/W2612445135","https://openalex.org/W2619516334","https://openalex.org/W2622263826","https://openalex.org/W2749988060","https://openalex.org/W2755543420","https://openalex.org/W2769856846","https://openalex.org/W2776855315","https://openalex.org/W2884711234","https://openalex.org/W2899771611","https://openalex.org/W2901541570","https://openalex.org/W2949650786","https://openalex.org/W2950179405","https://openalex.org/W2951781666","https://openalex.org/W2952020226","https://openalex.org/W2963470657","https://openalex.org/W2964121744","https://openalex.org/W2974168418","https://openalex.org/W3102816259","https://openalex.org/W3118608800","https://openalex.org/W3137695714"],"related_works":["https://openalex.org/W3162204513","https://openalex.org/W2371138613","https://openalex.org/W2048963458","https://openalex.org/W43109613","https://openalex.org/W2359952343","https://openalex.org/W2239445980","https://openalex.org/W2080152487","https://openalex.org/W4287755480","https://openalex.org/W2785875001","https://openalex.org/W4220659530"],"abstract_inverted_index":{"Current":[0],"large-scale":[1],"training":[2],"of":[3,19,56,77,85],"deep":[4],"neural":[5],"networks":[6],"typically":[7],"employs":[8],"synchronous":[9],"stochastic":[10],"gradient":[11],"descent":[12],"that":[13,31,41,53,64,93],"incurs":[14],"large":[15],"communication":[16],"overhead.":[17],"Instead":[18],"optimizing":[20],"reduction":[21,45,68,86,91],"routines":[22],"as":[23,48],"done":[24],"in":[25],"recent":[26],"studies,":[27],"we":[28,89],"propose":[29,61],"algorithms":[30],"do":[32],"not":[33],"require":[34],"frequent":[35],"global":[36,44,67],"reductions.":[37],"We":[38,59],"first":[39],"show":[40],"reducing":[42],"the":[43,66,83],"frequency":[46,69],"works":[47],"an":[49,62],"effective":[50],"regularization":[51],"technique":[52],"improves":[54],"generalization":[55],"adaptive":[57],"optimizers.":[58],"then":[60],"algorithm":[63],"reduces":[65],"by":[70],"employing":[71],"local":[72],"reductions":[73],"on":[74,87],"a":[75],"subset":[76],"learners.":[78],"In":[79],"addition,":[80],"to":[81],"maximize":[82],"effect":[84],"convergence,":[88],"introduce":[90],"momentum":[92],"further":[94],"accelerates":[95],"convergence.":[96]},"counts_by_year":[{"year":2021,"cited_by_count":1},{"year":2020,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}