{"id":"https://openalex.org/W7136390092","doi":"https://doi.org/10.1007/s44443-025-00452-1","title":"Accelerating dilated Winograd convolution with fused GPU kernel using tensor cores","display_name":"Accelerating dilated Winograd convolution with fused GPU kernel using tensor cores","publication_year":2026,"publication_date":"2026-03-16","ids":{"openalex":"https://openalex.org/W7136390092","doi":"https://doi.org/10.1007/s44443-025-00452-1"},"language":"en","primary_location":{"id":"doi:10.1007/s44443-025-00452-1","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s44443-025-00452-1","pdf_url":null,"source":{"id":"https://openalex.org/S2764955546","display_name":"Journal of King Saud University - Computer and Information Sciences","issn_l":"1319-1578","issn":["1319-1578","2213-1248"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of King Saud University Computer and Information Sciences","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1007/s44443-025-00452-1","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5033698712","display_name":"Xicheng Zhang","orcid":"https://orcid.org/0000-0003-3721-1357"},"institutions":[{"id":"https://openalex.org/I80143920","display_name":"Shandong University of Science and Technology","ror":"https://ror.org/04gtjhw98","country_code":"CN","type":"education","lineage":["https://openalex.org/I80143920"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shixiang Zhang","raw_affiliation_strings":["College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao, 266590, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao, 266590, China","institution_ids":["https://openalex.org/I80143920"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129591908","display_name":"Jianguo Liang","orcid":null},"institutions":[{"id":"https://openalex.org/I202126657","display_name":"Qufu Normal University","ror":"https://ror.org/03ceheh96","country_code":"CN","type":"education","lineage":["https://openalex.org/I202126657"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jianguo Liang","raw_affiliation_strings":["College of Computer, Qufu Normal University, Rizhao, 276826, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"College of Computer, Qufu Normal University, Rizhao, 276826, China","institution_ids":["https://openalex.org/I202126657"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085420086","display_name":"You Fu","orcid":"https://orcid.org/0000-0002-7809-4233"},"institutions":[{"id":"https://openalex.org/I80143920","display_name":"Shandong University of Science and Technology","ror":"https://ror.org/04gtjhw98","country_code":"CN","type":"education","lineage":["https://openalex.org/I80143920"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"You Fu","raw_affiliation_strings":["College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao, 266590, China"],"raw_orcid":"https://orcid.org/0000-0002-7809-4233","affiliations":[{"raw_affiliation_string":"College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao, 266590, China","institution_ids":["https://openalex.org/I80143920"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089872646","display_name":"Rong Hua","orcid":"https://orcid.org/0000-0002-1772-2419"},"institutions":[{"id":"https://openalex.org/I80143920","display_name":"Shandong University of Science and Technology","ror":"https://ror.org/04gtjhw98","country_code":"CN","type":"education","lineage":["https://openalex.org/I80143920"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Rong Hua","raw_affiliation_strings":["College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao, 266590, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao, 266590, China","institution_ids":["https://openalex.org/I80143920"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100576440","display_name":"Jianzhi Yu","orcid":"https://orcid.org/0000-0002-8024-6320"},"institutions":[{"id":"https://openalex.org/I80143920","display_name":"Shandong University of Science and Technology","ror":"https://ror.org/04gtjhw98","country_code":"CN","type":"education","lineage":["https://openalex.org/I80143920"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jianzhi Yu","raw_affiliation_strings":["College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao, 266590, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao, 266590, China","institution_ids":["https://openalex.org/I80143920"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100383897","display_name":"Qianqian Li","orcid":"https://orcid.org/0000-0002-3866-8159"},"institutions":[{"id":"https://openalex.org/I80143920","display_name":"Shandong University of Science and Technology","ror":"https://ror.org/04gtjhw98","country_code":"CN","type":"education","lineage":["https://openalex.org/I80143920"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qianqian Li","raw_affiliation_strings":["College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao, 266590, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao, 266590, China","institution_ids":["https://openalex.org/I80143920"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5085420086"],"corresponding_institution_ids":["https://openalex.org/I80143920"],"apc_list":{"value":1350,"currency":"USD","value_usd":1350},"apc_paid":{"value":1350,"currency":"USD","value_usd":1350},"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.25826635,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"38","issue":"5","first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.8830000162124634,"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":0.8830000162124634,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.010599999688565731,"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/T10719","display_name":"3D Shape Modeling and Analysis","score":0.008700000122189522,"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/convolution","display_name":"Convolution (computer science)","score":0.762499988079071},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.6773999929428101},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.5863999724388123},{"id":"https://openalex.org/keywords/convolutional-code","display_name":"Convolutional code","score":0.5712000131607056},{"id":"https://openalex.org/keywords/memory-footprint","display_name":"Memory footprint","score":0.48080000281333923},{"id":"https://openalex.org/keywords/tensor","display_name":"Tensor (intrinsic definition)","score":0.4699999988079071}],"concepts":[{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.762499988079071},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.6773999929428101},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6680999994277954},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.5863999724388123},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5738000273704529},{"id":"https://openalex.org/C157899210","wikidata":"https://www.wikidata.org/wiki/Q1395022","display_name":"Convolutional code","level":3,"score":0.5712000131607056},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.48080000281333923},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.4699999988079071},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4627000093460083},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.42730000615119934},{"id":"https://openalex.org/C181002996","wikidata":"https://www.wikidata.org/wiki/Q1611641","display_name":"Overlap\u2013add method","level":5,"score":0.37950000166893005},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3151000142097473},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3077999949455261},{"id":"https://openalex.org/C2780757906","wikidata":"https://www.wikidata.org/wiki/Q5276676","display_name":"Dilation (metric space)","level":2,"score":0.29010000824928284},{"id":"https://openalex.org/C164620267","wikidata":"https://www.wikidata.org/wiki/Q376953","display_name":"Adder","level":3,"score":0.2685999870300293}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1007/s44443-025-00452-1","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s44443-025-00452-1","pdf_url":null,"source":{"id":"https://openalex.org/S2764955546","display_name":"Journal of King Saud University - Computer and Information Sciences","issn_l":"1319-1578","issn":["1319-1578","2213-1248"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of King Saud University Computer and Information Sciences","raw_type":"journal-article"},{"id":"pmh:oai:doaj.org/article:50cd616ea7184d60b84bbd3b63d51cc7","is_oa":true,"landing_page_url":"https://doaj.org/article/50cd616ea7184d60b84bbd3b63d51cc7","pdf_url":null,"source":{"id":"https://openalex.org/S4306401280","display_name":"DOAJ (DOAJ: Directory of Open Access Journals)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Journal of King Saud University: Computer and Information Sciences, Vol 38, Iss 5 (2026)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.1007/s44443-025-00452-1","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s44443-025-00452-1","pdf_url":null,"source":{"id":"https://openalex.org/S2764955546","display_name":"Journal of King Saud University - Computer and Information Sciences","issn_l":"1319-1578","issn":["1319-1578","2213-1248"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of King Saud University Computer and Information Sciences","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1199161091","display_name":null,"funder_award_id":"ZR2025MS1098","funder_id":"https://openalex.org/F4320324174","funder_display_name":"Natural Science Foundation of Shandong Province"},{"id":"https://openalex.org/G77906310","display_name":null,"funder_award_id":"ZR2022MF274","funder_id":"https://openalex.org/F4320324174","funder_display_name":"Natural Science Foundation of Shandong Province"}],"funders":[{"id":"https://openalex.org/F4320324174","display_name":"Natural Science Foundation of Shandong Province","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":45,"referenced_works":["https://openalex.org/W1857884451","https://openalex.org/W2102605133","https://openalex.org/W2172654076","https://openalex.org/W2194775991","https://openalex.org/W2412782625","https://openalex.org/W2463631526","https://openalex.org/W2786374423","https://openalex.org/W2804199516","https://openalex.org/W2893801697","https://openalex.org/W2901549770","https://openalex.org/W2941408718","https://openalex.org/W2948661249","https://openalex.org/W2963989532","https://openalex.org/W2964209782","https://openalex.org/W2970938098","https://openalex.org/W2998957070","https://openalex.org/W3000160544","https://openalex.org/W3008378296","https://openalex.org/W3021029305","https://openalex.org/W3042016589","https://openalex.org/W3046538393","https://openalex.org/W3090600872","https://openalex.org/W3096609285","https://openalex.org/W3102337469","https://openalex.org/W3117765988","https://openalex.org/W3132532188","https://openalex.org/W3166036439","https://openalex.org/W3194710734","https://openalex.org/W3204140704","https://openalex.org/W4221160294","https://openalex.org/W4308090436","https://openalex.org/W4312603814","https://openalex.org/W4386764427","https://openalex.org/W4388739202","https://openalex.org/W4391387876","https://openalex.org/W4391958657","https://openalex.org/W4392884779","https://openalex.org/W4395106472","https://openalex.org/W4401408852","https://openalex.org/W4404417996","https://openalex.org/W4405755887","https://openalex.org/W4406164163","https://openalex.org/W4406548153","https://openalex.org/W4408029514","https://openalex.org/W7133219743"],"related_works":[],"abstract_inverted_index":{"Abstract":[0],"2D":[1,64],"convolutional":[2,145],"neural":[3],"networks":[4],"(CNNs)":[5],"achieve":[6],"remarkable":[7],"accuracy":[8],"across":[9,78],"diverse":[10],"computer":[11],"vision":[12],"tasks,":[13],"yet":[14],"their":[15],"inference":[16,219],"efficiency":[17],"remains":[18],"suboptimal.":[19],"As":[20],"a":[21,62,83,126],"fast":[22],"convolution":[23,26,30,49,66,76,94],"algorithm,":[24],"Winograd":[25,65,93],"can":[27],"significantly":[28],"accelerate":[29,111],"operations,":[31],"which":[32,72],"constitute":[33],"the":[34,90,112],"primary":[35],"performance":[36],"bottleneck":[37],"in":[38],"CNNs.":[39],"However,":[40],"existing":[41],"GPU-based":[42,63],"implementations":[43],"are":[44],"restricted":[45],"to":[46,110,165,187,209,242,264],"accelerating":[47],"standard":[48],"with":[50,203,279],"unit":[51],"dilation,":[52],"while":[53],"lacking":[54],"support":[55,283],"for":[56,69,143,284],"dilated":[57,70,75,96],"convolution.":[58,97],"We":[59],"propose":[60],"TC-DWC,":[61],"implementation":[67],"tailored":[68],"convolution,":[71],"effectively":[73],"accelerates":[74],"tasks":[77],"various":[79],"configurations.":[80],"TC-DWC":[81,147,150,205,212,224,227,281],"employs":[82],"two-step":[84],"tile":[85],"reorganization":[86],"scheme":[87],"that":[88,131],"reconciles":[89],"incompatibility":[91],"between":[92],"and":[95,148,176,213,222,225,253],"Moreover,":[98],"it":[99],"leverages":[100],"specialized":[101],"high-throughput":[102],"tensor":[103],"cores":[104],"(TCs),":[105],"replacing":[106],"conventional":[107],"vector":[108],"units":[109],"computationally":[113],"dominant":[114],"matrix":[115],"multiplications.":[116],"To":[117],"alleviate":[118],"TC-DWC\u2019s":[119],"substantial":[120],"memory":[121,207],"footprint,":[122],"we":[123],"further":[124],"develop":[125],"multi-stage":[127],"kernel":[128],"fusion":[129],"strategy":[130],"eliminates":[132],"intermediate":[133],"array":[134],"allocations,":[135],"yielding":[136],"Fused":[137,149,204,226,280],"TC-DWC.":[138],"Experimental":[139],"results":[140],"demonstrate":[141],"that,":[142],"single":[144],"layers,":[146],"deliver":[151,228],"average":[152,229],"speedups":[153,230],"of":[154,211,215,220,231],"$$1.56\\times":[155],"$$":[156,167,178,189,233,244,255,266],"<mml:math":[157,168,179,190,234,245,256,267],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\">":[158,169,180,191,235,246,257,268],"<mml:mrow>":[159,170,181,192,236,247,258,269],"<mml:mn>1.56</mml:mn>":[160],"<mml:mo>\u00d7</mml:mo>":[161,172,183,194,238,249,260,271],"</mml:mrow>":[162,173,184,195,239,250,261,272],"</mml:math>":[163,174,185,196,240,251,262,273],"(up":[164,186,241,263],"$$2.45\\times":[166],"<mml:mn>2.45</mml:mn>":[171],")":[175,197,252,274],"$$1.29\\times":[177],"<mml:mn>1.29</mml:mn>":[182],"$$1.80\\times":[188],"<mml:mn>1.80</mml:mn>":[193],"over":[198,275],"cuDNN":[199],"optimal":[200],"implementation,":[201],"respectively,":[202,278],"reducing":[206],"footprint":[208],"35%":[210],"55%":[214],"cuDNN.":[216],"For":[217],"end-to-end":[218],"CSRNet":[221],"DLinkNet34,":[223],"$$1.38\\times":[232],"<mml:mn>1.38</mml:mn>":[237],"$$1.41\\times":[243],"<mml:mn>1.41</mml:mn>":[248],"$$1.20\\times":[254],"<mml:mn>1.20</mml:mn>":[259],"$$1.22\\times":[265],"<mml:mn>1.22</mml:mn>":[270],"PyTorch":[276],"baseline,":[277],"enabling":[282],"larger":[285],"batch":[286],"sizes.":[287]},"counts_by_year":[],"updated_date":"2026-06-13T06:13:01.061226","created_date":"2026-03-17T00:00:00"}