{"id":"https://openalex.org/W4221113823","doi":"https://doi.org/10.1109/tpds.2022.3161187","title":"Deep Neural Network Training With Distributed K-FAC","display_name":"Deep Neural Network Training With Distributed K-FAC","publication_year":2022,"publication_date":"2022-03-22","ids":{"openalex":"https://openalex.org/W4221113823","doi":"https://doi.org/10.1109/tpds.2022.3161187"},"language":"en","primary_location":{"id":"doi:10.1109/tpds.2022.3161187","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2022.3161187","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Parallel and Distributed Systems","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/A5069913075","display_name":"J. Gregory Pauloski","orcid":"https://orcid.org/0000-0002-6547-6902"},"institutions":[{"id":"https://openalex.org/I40347166","display_name":"University of Chicago","ror":"https://ror.org/024mw5h28","country_code":"US","type":"education","lineage":["https://openalex.org/I40347166"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"J. Gregory Pauloski","raw_affiliation_strings":["Department of Computer Science, University of Chicago, Chicago, IL, USA"],"raw_orcid":"https://orcid.org/0000-0002-6547-6902","affiliations":[{"raw_affiliation_string":"Department of Computer Science, University of Chicago, Chicago, IL, USA","institution_ids":["https://openalex.org/I40347166"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101670806","display_name":"Lei Huang","orcid":"https://orcid.org/0000-0003-0502-168X"},"institutions":[{"id":"https://openalex.org/I4388891828","display_name":"Texas Advanced Computing Center","ror":"https://ror.org/00xg4bh43","country_code":null,"type":"facility","lineage":["https://openalex.org/I4388891828","https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Lei Huang","raw_affiliation_strings":["Texas Advanced Computing Center, Austin, TX, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Texas Advanced Computing Center, Austin, TX, USA","institution_ids":["https://openalex.org/I4388891828"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101883722","display_name":"Weijia Xu","orcid":"https://orcid.org/0000-0002-5134-6381"},"institutions":[{"id":"https://openalex.org/I4388891828","display_name":"Texas Advanced Computing Center","ror":"https://ror.org/00xg4bh43","country_code":null,"type":"facility","lineage":["https://openalex.org/I4388891828","https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Weijia Xu","raw_affiliation_strings":["Texas Advanced Computing Center, Austin, TX, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Texas Advanced Computing Center, Austin, TX, USA","institution_ids":["https://openalex.org/I4388891828"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065464552","display_name":"Kyle Chard","orcid":"https://orcid.org/0000-0002-7370-4805"},"institutions":[{"id":"https://openalex.org/I40347166","display_name":"University of Chicago","ror":"https://ror.org/024mw5h28","country_code":"US","type":"education","lineage":["https://openalex.org/I40347166"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kyle Chard","raw_affiliation_strings":["Department of Computer Science, University of Chicago, Chicago, IL, USA"],"raw_orcid":"https://orcid.org/0000-0002-7370-4805","affiliations":[{"raw_affiliation_string":"Department of Computer Science, University of Chicago, Chicago, IL, USA","institution_ids":["https://openalex.org/I40347166"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032231503","display_name":"Ian Foster","orcid":"https://orcid.org/0000-0003-2129-5269"},"institutions":[{"id":"https://openalex.org/I40347166","display_name":"University of Chicago","ror":"https://ror.org/024mw5h28","country_code":"US","type":"education","lineage":["https://openalex.org/I40347166"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ian T. Foster","raw_affiliation_strings":["Department of Computer Science, University of Chicago, Chicago, IL, USA"],"raw_orcid":"https://orcid.org/0000-0003-2129-5269","affiliations":[{"raw_affiliation_string":"Department of Computer Science, University of Chicago, Chicago, IL, USA","institution_ids":["https://openalex.org/I40347166"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100422982","display_name":"Zhao Zhang","orcid":"https://orcid.org/0000-0001-5921-0035"},"institutions":[{"id":"https://openalex.org/I4388891828","display_name":"Texas Advanced Computing Center","ror":"https://ror.org/00xg4bh43","country_code":null,"type":"facility","lineage":["https://openalex.org/I4388891828","https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhao Zhang","raw_affiliation_strings":["Texas Advanced Computing Center, Austin, TX, USA"],"raw_orcid":"https://orcid.org/0000-0001-5921-0035","affiliations":[{"raw_affiliation_string":"Texas Advanced Computing Center, Austin, TX, USA","institution_ids":["https://openalex.org/I4388891828"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.2936,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.50673685,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":97},"biblio":{"volume":"33","issue":"12","first_page":"3616","last_page":"3627"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9984999895095825,"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":0.9984999895095825,"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/T12549","display_name":"Image and Object Detection Techniques","score":0.9919000267982483,"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/T10320","display_name":"Neural Networks and Applications","score":0.9908000230789185,"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":0.8075559139251709},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6390666961669922},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.5890710353851318},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.47404715418815613},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.342604398727417}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8075559139251709},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6390666961669922},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5890710353851318},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47404715418815613},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.342604398727417},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tpds.2022.3161187","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2022.3161187","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Parallel and Distributed Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1408889976","display_name":null,"funder_award_id":"OAC-1931354","funder_id":"https://openalex.org/F4320335353","funder_display_name":"National Science Foundation of Sri Lanka"},{"id":"https://openalex.org/G1713112429","display_name":null,"funder_award_id":"OAC-2106661","funder_id":"https://openalex.org/F4320335353","funder_display_name":"National Science Foundation of Sri Lanka"},{"id":"https://openalex.org/G768599749","display_name":null,"funder_award_id":"OAC-1818253","funder_id":"https://openalex.org/F4320335353","funder_display_name":"National Science Foundation of Sri Lanka"},{"id":"https://openalex.org/G961797300","display_name":null,"funder_award_id":"OAC-2107511","funder_id":"https://openalex.org/F4320335353","funder_display_name":"National Science Foundation of Sri Lanka"}],"funders":[{"id":"https://openalex.org/F4320306084","display_name":"U.S. Department of Energy","ror":"https://ror.org/01bj3aw27"},{"id":"https://openalex.org/F4320335353","display_name":"National Science Foundation of Sri Lanka","ror":"https://ror.org/010xaa060"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":74,"referenced_works":["https://openalex.org/W1566289585","https://openalex.org/W1861492603","https://openalex.org/W2051434435","https://openalex.org/W2060393849","https://openalex.org/W2108598243","https://openalex.org/W2131613942","https://openalex.org/W2138243089","https://openalex.org/W2155894447","https://openalex.org/W2186615578","https://openalex.org/W2194775991","https://openalex.org/W2337082154","https://openalex.org/W2622263826","https://openalex.org/W2750933313","https://openalex.org/W2757910899","https://openalex.org/W2763421725","https://openalex.org/W2769856846","https://openalex.org/W2786263712","https://openalex.org/W2806659157","https://openalex.org/W2896457183","https://openalex.org/W2901541570","https://openalex.org/W2903697572","https://openalex.org/W2920668770","https://openalex.org/W2937394206","https://openalex.org/W2953384591","https://openalex.org/W2954650510","https://openalex.org/W2962747323","https://openalex.org/W2963150697","https://openalex.org/W2963408258","https://openalex.org/W2963433607","https://openalex.org/W2963792515","https://openalex.org/W2963804082","https://openalex.org/W2963959597","https://openalex.org/W2967791890","https://openalex.org/W2988070836","https://openalex.org/W2991040477","https://openalex.org/W2997905860","https://openalex.org/W2998028735","https://openalex.org/W3037045207","https://openalex.org/W3081168214","https://openalex.org/W3081233814","https://openalex.org/W3118608800","https://openalex.org/W3131087743","https://openalex.org/W3180201702","https://openalex.org/W4294641903","https://openalex.org/W4295181861","https://openalex.org/W4295312788","https://openalex.org/W4298840580","https://openalex.org/W4301239768","https://openalex.org/W6638803421","https://openalex.org/W6680402377","https://openalex.org/W6683107984","https://openalex.org/W6686509673","https://openalex.org/W6693969136","https://openalex.org/W6713134421","https://openalex.org/W6726983090","https://openalex.org/W6730009442","https://openalex.org/W6739622702","https://openalex.org/W6743885270","https://openalex.org/W6744513255","https://openalex.org/W6745245109","https://openalex.org/W6746419290","https://openalex.org/W6746514494","https://openalex.org/W6748315838","https://openalex.org/W6748645090","https://openalex.org/W6752294818","https://openalex.org/W6755207826","https://openalex.org/W6756439839","https://openalex.org/W6756718674","https://openalex.org/W6757053730","https://openalex.org/W6759786419","https://openalex.org/W6760600040","https://openalex.org/W6763067651","https://openalex.org/W6766978945","https://openalex.org/W6787972765"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W230091440","https://openalex.org/W2390279801","https://openalex.org/W2233261550","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2810751659","https://openalex.org/W258997015","https://openalex.org/W2376932109"],"abstract_inverted_index":{"Scaling":[0],"deep":[1,93],"neural":[2,94],"network":[3,95],"training":[4,17,33,120],"to":[5,14,45,71,149],"more":[6],"processors":[7],"and":[8,23,49,87,107,126,139],"larger":[9,27],"batch":[10],"sizes":[11],"is":[12,29],"key":[13],"reducing":[15],"end-to-end":[16],"time;":[18],"yet,":[19],"maintaining":[20],"comparable":[21],"convergence":[22,117,125],"hardware":[24],"utilization":[25],"at":[26],"scales":[28,34],"challenging.":[30],"Increases":[31],"in":[32,91,153],"have":[35],"enabled":[36],"natural":[37,57],"gradient":[38,47,58,105,132],"optimization":[39],"methods":[40],"as":[41],"a":[42,56,83,166],"reasonable":[43],"alternative":[44],"stochastic":[46],"descent":[48],"variants":[50],"thereof.":[51],"Kronecker-factored":[52],"Approximate":[53],"Curvature":[54],"(K-FAC),":[55],"method,":[59],"preconditions":[60],"gradients":[61],"with":[62,112],"an":[63,77],"efficient":[64],"approximation":[65],"of":[66,115,129,168],"the":[67,113,124,158],"Fisher":[68],"Information":[69],"Matrix":[70],"improve":[72],"per-iteration":[73],"progress":[74],"when":[75],"optimizing":[76],"objective":[78],"function.":[79],"Here":[80],"we":[81,98],"propose":[82],"scalable":[84],"K-FAC":[85,109,131],"algorithm":[86],"investigate":[88],"K-FAC\u2019s":[89],"applicability":[90],"large-scale":[92],"training.":[96],"Specifically,":[97],"explore":[99],"layer-wise":[100],"distribution":[101],"strategies,":[102],"inverse-free":[103],"second-order":[104],"evaluation,":[106],"dynamic":[108],"update":[110],"decoupling,":[111],"goal":[114],"preserving":[116],"while":[118],"minimizing":[119],"time.":[121],"We":[122],"evaluate":[123],"scaling":[127],"properties":[128],"our":[130,146],"preconditioner,":[133],"for":[134],"image":[135],"classification,":[136],"object":[137],"detection,":[138],"language":[140],"modeling":[141],"applications.":[142],"In":[143],"all":[144],"applications,":[145],"implementation":[147],"converges":[148],"baseline":[150],"performance":[151],"targets":[152],"9\u201325%":[154],"less":[155],"time":[156],"than":[157],"standard":[159],"first-order":[160],"optimizers":[161],"on":[162],"GPU":[163],"clusters":[164],"across":[165],"variety":[167],"scales.":[169]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}