{"id":"https://openalex.org/W2736396799","doi":"https://doi.org/10.1109/tpds.2017.2731764","title":"A GPU-Architecture Optimized Hierarchical Decomposition Algorithm for Support Vector Machine Training","display_name":"A GPU-Architecture Optimized Hierarchical Decomposition Algorithm for Support Vector Machine Training","publication_year":2017,"publication_date":"2017-07-25","ids":{"openalex":"https://openalex.org/W2736396799","doi":"https://doi.org/10.1109/tpds.2017.2731764","mag":"2736396799"},"language":"en","primary_location":{"id":"doi:10.1109/tpds.2017.2731764","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2017.2731764","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/A5067707474","display_name":"Jan Van\u011bk","orcid":"https://orcid.org/0000-0002-2639-6731"},"institutions":[{"id":"https://openalex.org/I92715842","display_name":"University of West Bohemia","ror":"https://ror.org/040t43x18","country_code":"CZ","type":"education","lineage":["https://openalex.org/I92715842"]}],"countries":["CZ"],"is_corresponding":true,"raw_author_name":"Jan Vanek","raw_affiliation_strings":["University of West Bohemia, New Technologies for the Information Society, Pilsen, Czech Republic"],"affiliations":[{"raw_affiliation_string":"University of West Bohemia, New Technologies for the Information Society, Pilsen, Czech Republic","institution_ids":["https://openalex.org/I92715842"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013751340","display_name":"Josef Mich\u00e1lek","orcid":"https://orcid.org/0000-0001-7757-3163"},"institutions":[{"id":"https://openalex.org/I92715842","display_name":"University of West Bohemia","ror":"https://ror.org/040t43x18","country_code":"CZ","type":"education","lineage":["https://openalex.org/I92715842"]}],"countries":["CZ"],"is_corresponding":false,"raw_author_name":"Josef Michalek","raw_affiliation_strings":["University of West Bohemia, New Technologies for the Information Society, Pilsen, Czech Republic"],"affiliations":[{"raw_affiliation_string":"University of West Bohemia, New Technologies for the Information Society, Pilsen, Czech Republic","institution_ids":["https://openalex.org/I92715842"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5103070222","display_name":"Josef Psutka","orcid":"https://orcid.org/0000-0002-0764-3207"},"institutions":[{"id":"https://openalex.org/I92715842","display_name":"University of West Bohemia","ror":"https://ror.org/040t43x18","country_code":"CZ","type":"education","lineage":["https://openalex.org/I92715842"]}],"countries":["CZ"],"is_corresponding":false,"raw_author_name":"Josef Psutka","raw_affiliation_strings":["University of West Bohemia, New Technologies for the Information Society, Pilsen, Czech Republic"],"affiliations":[{"raw_affiliation_string":"University of West Bohemia, New Technologies for the Information Society, Pilsen, Czech Republic","institution_ids":["https://openalex.org/I92715842"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5067707474"],"corresponding_institution_ids":["https://openalex.org/I92715842"],"apc_list":null,"apc_paid":null,"fwci":1.0012,"has_fulltext":false,"cited_by_count":20,"citation_normalized_percentile":{"value":0.85068793,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":98},"biblio":{"volume":"28","issue":"12","first_page":"3330","last_page":"3343"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10812","display_name":"Human Pose and Action Recognition","score":0.9939000010490417,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9939000010490417,"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/T12303","display_name":"Tensor decomposition and applications","score":0.992900013923645,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10500","display_name":"Sparse and Compressive Sensing Techniques","score":0.9897000193595886,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8431069254875183},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.6408722400665283},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6330921649932861},{"id":"https://openalex.org/keywords/support-vector-machine","display_name":"Support vector machine","score":0.5763601064682007},{"id":"https://openalex.org/keywords/solver","display_name":"Solver","score":0.5337582230567932},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.5137442350387573},{"id":"https://openalex.org/keywords/sequential-minimal-optimization","display_name":"Sequential minimal optimization","score":0.5100951790809631},{"id":"https://openalex.org/keywords/matrix-decomposition","display_name":"Matrix decomposition","score":0.43851232528686523},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.19303041696548462},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.10592275857925415}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8431069254875183},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.6408722400665283},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6330921649932861},{"id":"https://openalex.org/C12267149","wikidata":"https://www.wikidata.org/wiki/Q282453","display_name":"Support vector machine","level":2,"score":0.5763601064682007},{"id":"https://openalex.org/C2778770139","wikidata":"https://www.wikidata.org/wiki/Q1966904","display_name":"Solver","level":2,"score":0.5337582230567932},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5137442350387573},{"id":"https://openalex.org/C10719679","wikidata":"https://www.wikidata.org/wiki/Q2679145","display_name":"Sequential minimal optimization","level":3,"score":0.5100951790809631},{"id":"https://openalex.org/C42355184","wikidata":"https://www.wikidata.org/wiki/Q1361088","display_name":"Matrix decomposition","level":3,"score":0.43851232528686523},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.19303041696548462},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.10592275857925415},{"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/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","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/C158693339","wikidata":"https://www.wikidata.org/wiki/Q190524","display_name":"Eigenvalues and eigenvectors","level":2,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tpds.2017.2731764","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2017.2731764","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":[{"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure","score":0.44999998807907104}],"awards":[{"id":"https://openalex.org/G8397218950","display_name":null,"funder_award_id":"GBP103/12/G084","funder_id":"https://openalex.org/F4320321006","funder_display_name":"Grantov\u00e1 Agentura \u010cesk\u00e9 Republiky"}],"funders":[{"id":"https://openalex.org/F4320321006","display_name":"Grantov\u00e1 Agentura \u010cesk\u00e9 Republiky","ror":"https://ror.org/01pv73b02"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":49,"referenced_works":["https://openalex.org/W127032682","https://openalex.org/W144150909","https://openalex.org/W1171471126","https://openalex.org/W1506135481","https://openalex.org/W1512098439","https://openalex.org/W1576520375","https://openalex.org/W1788526377","https://openalex.org/W1975713049","https://openalex.org/W1989565472","https://openalex.org/W1993235091","https://openalex.org/W2025208635","https://openalex.org/W2032148848","https://openalex.org/W2035720976","https://openalex.org/W2064517612","https://openalex.org/W2065887373","https://openalex.org/W2081621367","https://openalex.org/W2087624898","https://openalex.org/W2088730795","https://openalex.org/W2095885101","https://openalex.org/W2100522629","https://openalex.org/W2110122118","https://openalex.org/W2115694019","https://openalex.org/W2116173306","https://openalex.org/W2117990954","https://openalex.org/W2118075434","https://openalex.org/W2122565017","https://openalex.org/W2132425109","https://openalex.org/W2150062669","https://openalex.org/W2151040995","https://openalex.org/W2151911322","https://openalex.org/W2159350554","https://openalex.org/W2161238936","https://openalex.org/W2165966284","https://openalex.org/W2170860445","https://openalex.org/W2241715985","https://openalex.org/W2489902977","https://openalex.org/W2514172024","https://openalex.org/W2593370983","https://openalex.org/W2738579668","https://openalex.org/W3007294223","https://openalex.org/W4285719527","https://openalex.org/W6630527977","https://openalex.org/W6634442568","https://openalex.org/W6638167705","https://openalex.org/W6677380243","https://openalex.org/W6677732001","https://openalex.org/W6677987841","https://openalex.org/W6682347367","https://openalex.org/W6741128498"],"related_works":["https://openalex.org/W1592900392","https://openalex.org/W2163526212","https://openalex.org/W2036133907","https://openalex.org/W2394447189","https://openalex.org/W2539940072","https://openalex.org/W2092829304","https://openalex.org/W2538830880","https://openalex.org/W2366786074","https://openalex.org/W2090763504","https://openalex.org/W2364362706"],"abstract_inverted_index":{"In":[0,84,248],"the":[1,38,57,94,218,233,251,270,275,292,303,317,327,345],"last":[2],"decade,":[3],"several":[4],"GPU":[5,91,117,126,137,246,321],"implementations":[6],"of":[7,19,45,59,68,80,173,230,239,269,282],"Support":[8],"Vector":[9],"Machine":[10],"(SVM)":[11],"training":[12,96,175,245],"with":[13,22,163,227,316,340],"nonlinear":[14],"kernels":[15],"were":[16],"published.":[17],"Some":[18],"them":[20],"even":[21],"source":[23],"codes.":[24],"The":[25,119,167,184,298],"most":[26,268],"effective":[27],"ones":[28],"are":[29,50,146,192,259],"based":[30],"on":[31,76,278,302],"Sequential":[32],"Minimal":[33],"Optimization":[34],"(SMO).":[35],"They":[36],"decompose":[37],"restricted":[39],"quadratic":[40],"problem":[41],"into":[42,261],"a":[43,65,89,107,124,131,135,157,179,198,209,228,279],"series":[44],"smallest":[46],"possible":[47],"subproblems,":[48],"which":[49],"then":[51],"solved":[52],"analytically.":[53],"For":[54],"large":[55,66,199],"datasets,":[56,285],"majority":[58],"elapsed":[60],"time":[61],"is":[62,152,213,265,326],"spent":[63],"by":[64],"amount":[67],"matrix-vector":[69,219],"multiplications":[70],"that":[71,97,112,145,191,212,330],"cannot":[72],"be":[73],"computed":[74],"efficiently":[75,129],"current":[77],"GPUs":[78],"because":[79],"limited":[81],"memory":[82],"bandwidth.":[83],"this":[85,249,341],"paper,":[86,250,342],"we":[87,98,307,343],"introduce":[88],"novel":[90],"approach":[92,288],"to":[93,115,128,148,194,274,311],"SVM":[95,103,244],"call":[99],"Optimized":[100],"Hierarchical":[101],"Decomposition":[102],"(OHD-SVM).":[104],"It":[105,151],"uses":[106,123],"hierarchical":[108],"decomposition":[109,121,169],"iterative":[110],"algorithm":[111],"fits":[113],"better":[114],"actual":[116],"architecture.":[118],"low":[120],"level":[122,170],"single":[125,136,158],"multiprocessor":[127,138],"solve":[130],"local":[132,161,165,181,200],"subproblem.":[133],"Nowadays":[134],"can":[139,331],"run":[140],"thousand":[141],"or":[142],"more":[143,215],"threads":[144],"able":[147],"synchronize":[149],"quickly.":[150],"an":[153,203,236],"ideal":[154],"platform":[155],"for":[156],"kernel":[159,189,205,257],"SMO-based":[160],"solver":[162],"fast":[164],"iterations.":[166],"high":[168],"updates":[171],"gradients":[172],"entire":[174],"set":[176,281],"and":[177,255],"selects":[178],"new":[180],"working":[182],"set.":[183],"gradient":[185],"update":[186],"requires":[187],"many":[188],"values":[190,206],"costly":[193],"compute.":[195],"However,":[196],"solving":[197],"subproblem":[201],"offers":[202],"efficient":[204,216],"computation":[207],"via":[208],"matrix-matrix":[210],"multiplication":[211,220],"much":[214],"than":[217],"used":[221],"in":[222,267,295,314],"already":[223,319],"published":[224,320,344],"implementations.":[225,247],"Along":[226,339],"description":[229],"our":[231,286,324],"implementation,":[232],"paper":[234],"includes":[235],"exact":[237],"comparison":[238,315],"five":[240],"publicly":[241,283],"available":[242,284],"C++":[243],"binary":[252],"classification":[253],"task":[254],"RBF":[256],"function":[258],"taken":[260],"account":[262],"as":[263,334,336],"it":[264],"usual":[266],"recent":[271],"papers.":[272],"According":[273],"measured":[276],"results":[277],"wide":[280],"proposed":[287],"excelled":[289],"significantly":[290],"over":[291],"other":[293],"methods":[294],"all":[296],"datasets.":[297,338],"biggest":[299],"difference":[300],"was":[301],"largest":[304],"dataset":[305],"where":[306],"achieved":[308],"speed-up":[309],"up":[310],"12":[312],"times":[313],"fastest":[318],"implementation.":[322],"Moreover,":[323],"OHD-SVM":[325],"only":[328],"one":[329],"handle":[332],"dense":[333],"well":[335],"sparse":[337],"source-codes":[346],"at":[347],"https://github.com/OrcusCZ/OHD-SVM.":[348]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2023,"cited_by_count":4},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":3},{"year":2020,"cited_by_count":2},{"year":2019,"cited_by_count":4},{"year":2018,"cited_by_count":5}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}