{"id":"https://openalex.org/W2765865054","doi":"https://doi.org/10.1145/3146347.3146351","title":"Accelerating deep neural network learning for speech recognition on a cluster of GPUs","display_name":"Accelerating deep neural network learning for speech recognition on a cluster of GPUs","publication_year":2017,"publication_date":"2017-10-31","ids":{"openalex":"https://openalex.org/W2765865054","doi":"https://doi.org/10.1145/3146347.3146351","mag":"2765865054"},"language":"en","primary_location":{"id":"doi:10.1145/3146347.3146351","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3146347.3146351","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Machine Learning on HPC Environments","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/A5089801879","display_name":"Guojing Cong","orcid":"https://orcid.org/0000-0003-0850-7714"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guojing Cong","raw_affiliation_strings":["IBM Research, Yorktown Heights, New York"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research, Yorktown Heights, New York","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003725957","display_name":"Brian Kingsbury","orcid":"https://orcid.org/0000-0002-1343-6837"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Brian Kingsbury","raw_affiliation_strings":["IBM Research, Yorktown Heights, New York"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research, Yorktown Heights, New York","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056519750","display_name":"Soumyadip Gosh","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Soumyadip Gosh","raw_affiliation_strings":["IBM Research, Yorktown Heights, New York"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research, Yorktown Heights, New York","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079994647","display_name":"George Saon","orcid":"https://orcid.org/0009-0004-6837-5009"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"George Saon","raw_affiliation_strings":["IBM Research, Yorktown Heights, New York"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research, Yorktown Heights, New York","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5103471288","display_name":"Fan Zhou","orcid":null},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Fan Zhou","raw_affiliation_strings":["Georgia Institute of Technology, North Avenue NW, Atlanta, Georgia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Georgia Institute of Technology, North Avenue NW, Atlanta, Georgia","institution_ids":["https://openalex.org/I130701444"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9994999766349792,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9994999766349792,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9983000159263611,"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/T11309","display_name":"Music and Audio Processing","score":0.9943000078201294,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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":0.8771129846572876},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.855502188205719},{"id":"https://openalex.org/keywords/gpu-cluster","display_name":"GPU cluster","score":0.7514695525169373},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.7202831506729126},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.6341889500617981},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5951704382896423},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5098225474357605},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.41575175523757935},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4023357033729553},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.26248443126678467},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.09162986278533936}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8771129846572876},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.855502188205719},{"id":"https://openalex.org/C2781335571","wikidata":"https://www.wikidata.org/wiki/Q2633544","display_name":"GPU cluster","level":3,"score":0.7514695525169373},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.7202831506729126},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.6341889500617981},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5951704382896423},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5098225474357605},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.41575175523757935},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4023357033729553},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.26248443126678467},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.09162986278533936}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3146347.3146351","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3146347.3146351","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Machine Learning on HPC Environments","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W778657980","https://openalex.org/W1522301498","https://openalex.org/W2058641082","https://openalex.org/W2060393849","https://openalex.org/W2061570747","https://openalex.org/W2064675550","https://openalex.org/W2109125156","https://openalex.org/W2130062883","https://openalex.org/W2137644731","https://openalex.org/W2146502635","https://openalex.org/W2158069733","https://openalex.org/W2166637769","https://openalex.org/W2168231600","https://openalex.org/W2288502450","https://openalex.org/W2559655401","https://openalex.org/W2604272474","https://openalex.org/W2783919261","https://openalex.org/W2883109957","https://openalex.org/W2951781666","https://openalex.org/W2963433607","https://openalex.org/W2963804082","https://openalex.org/W2964121744","https://openalex.org/W2998508934","https://openalex.org/W3102816259","https://openalex.org/W4245104967"],"related_works":["https://openalex.org/W1966837078","https://openalex.org/W4377865163","https://openalex.org/W3193857078","https://openalex.org/W2888956734","https://openalex.org/W3000197790","https://openalex.org/W4315865067","https://openalex.org/W2979433843","https://openalex.org/W3208304128","https://openalex.org/W2514049505","https://openalex.org/W1993048673"],"abstract_inverted_index":{"We":[0,16,46,73,95],"train":[1],"deep":[2],"neural":[3],"networks":[4],"to":[5,35,88,106],"solve":[6],"the":[7,28,63,68,92,107,121],"acoustic":[8],"modeling":[9],"problem":[10],"for":[11,51,56],"large-vocabulary":[12],"continuous":[13],"speech":[14],"recognition.":[15],"employ":[17],"distributed":[18],"processing":[19],"using":[20],"a":[21,33,75,125],"cluster":[22,109],"of":[23,110],"GPUs.":[24,112],"On":[25],"modern":[26],"GPUs,":[27,62],"sequential":[29,93,122],"implementation":[30,104,116,123],"takes":[31],"over":[32],"day":[34],"train,":[36],"and":[37,67],"efficient":[38,55],"parallelization":[39,52],"without":[40],"losing":[41],"accuracy":[42],"is":[43,65],"notoriously":[44],"hard.":[45],"show":[47],"that":[48,81,89,100],"ASGD":[49],"methods":[50],"are":[53,71],"not":[54],"this":[57],"application.":[58],"Even":[59],"with":[60,82,124],"4":[61,83],"overhead":[64],"significant,":[66],"accuracies":[69,86,119],"achieved":[70,90],"poor.":[72],"adapt":[74],"P-learner":[76],"K-step":[77],"model":[78],"averaging":[79],"algorithm":[80],"GPUs":[84],"achieves":[85,117],"comparable":[87],"by":[91],"implementation.":[94],"further":[96],"introduce":[97],"adaptive":[98],"measures":[99],"make":[101],"our":[102,114],"parallel":[103,115],"scale":[105],"full":[108],"20":[111],"Ultimately":[113],"better":[118],"than":[120],"6.1":[126],"times":[127],"speedup.":[128]},"counts_by_year":[{"year":2020,"cited_by_count":1},{"year":2019,"cited_by_count":1},{"year":2018,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}