{"id":"https://openalex.org/W7155156409","doi":"https://doi.org/10.1109/access.2026.3686628","title":"Fast GPU Inference With Unstructured-Pruned Sparse DNN Architectures","display_name":"Fast GPU Inference With Unstructured-Pruned Sparse DNN Architectures","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7155156409","doi":"https://doi.org/10.1109/access.2026.3686628"},"language":"en","primary_location":{"id":"doi:10.1109/access.2026.3686628","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2026.3686628","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1109/access.2026.3686628","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5070148027","display_name":"M. Ando","orcid":"https://orcid.org/0000-0003-3247-9565"},"institutions":[{"id":"https://openalex.org/I65143321","display_name":"Hitachi (Japan)","ror":"https://ror.org/02exqgm79","country_code":"JP","type":"company","lineage":["https://openalex.org/I65143321"]}],"countries":["JP"],"is_corresponding":true,"raw_author_name":"Masahiko Ando","raw_affiliation_strings":["Research and Development Group, Hitachi, Ltd., Kokubunji, Tokyo, Japan"],"raw_orcid":"https://orcid.org/0000-0003-3247-9565","affiliations":[{"raw_affiliation_string":"Research and Development Group, Hitachi, Ltd., Kokubunji, Tokyo, Japan","institution_ids":["https://openalex.org/I65143321"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008995588","display_name":"Keita Yamane","orcid":"https://orcid.org/0009-0004-2382-6670"},"institutions":[{"id":"https://openalex.org/I65143321","display_name":"Hitachi (Japan)","ror":"https://ror.org/02exqgm79","country_code":"JP","type":"company","lineage":["https://openalex.org/I65143321"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Keita Yamane","raw_affiliation_strings":["Research and Development Group, Hitachi, Ltd., Kokubunji, Tokyo, Japan"],"raw_orcid":"https://orcid.org/0009-0004-2382-6670","affiliations":[{"raw_affiliation_string":"Research and Development Group, Hitachi, Ltd., Kokubunji, Tokyo, Japan","institution_ids":["https://openalex.org/I65143321"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5036668422","display_name":"T. Oshima","orcid":"https://orcid.org/0000-0002-1069-3221"},"institutions":[{"id":"https://openalex.org/I65143321","display_name":"Hitachi (Japan)","ror":"https://ror.org/02exqgm79","country_code":"JP","type":"company","lineage":["https://openalex.org/I65143321"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Takashi Oshima","raw_affiliation_strings":["Research and Development Group, Hitachi, Ltd., Kokubunji, Tokyo, Japan"],"raw_orcid":"https://orcid.org/0000-0002-1069-3221","affiliations":[{"raw_affiliation_string":"Research and Development Group, Hitachi, Ltd., Kokubunji, Tokyo, Japan","institution_ids":["https://openalex.org/I65143321"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5070148027"],"corresponding_institution_ids":["https://openalex.org/I65143321"],"apc_list":{"value":1850,"currency":"USD","value_usd":1850},"apc_paid":{"value":1850,"currency":"USD","value_usd":1850},"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.77247254,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"14","issue":null,"first_page":"68271","last_page":"68280"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.7590000033378601,"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":0.7590000033378601,"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":0.025499999523162842,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.02160000056028366,"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/inference","display_name":"Inference","score":0.5224000215530396},{"id":"https://openalex.org/keywords/sparse-matrix","display_name":"Sparse matrix","score":0.33390000462532043},{"id":"https://openalex.org/keywords/sparse-approximation","display_name":"Sparse approximation","score":0.3188000023365021},{"id":"https://openalex.org/keywords/graphics-processing-unit","display_name":"Graphics processing unit","score":0.29919999837875366},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.2955000102519989},{"id":"https://openalex.org/keywords/approximate-inference","display_name":"Approximate inference","score":0.2948000133037567}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8345000147819519},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5224000215530396},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4975000023841858},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.33390000462532043},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3249000012874603},{"id":"https://openalex.org/C124066611","wikidata":"https://www.wikidata.org/wiki/Q28684319","display_name":"Sparse approximation","level":2,"score":0.3188000023365021},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3127000033855438},{"id":"https://openalex.org/C2779851693","wikidata":"https://www.wikidata.org/wiki/Q183484","display_name":"Graphics processing unit","level":2,"score":0.29919999837875366},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2955000102519989},{"id":"https://openalex.org/C2777472644","wikidata":"https://www.wikidata.org/wiki/Q16968992","display_name":"Approximate inference","level":3,"score":0.2948000133037567},{"id":"https://openalex.org/C104267543","wikidata":"https://www.wikidata.org/wiki/Q208163","display_name":"Signal processing","level":3,"score":0.29280000925064087},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.2879999876022339},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.27720001339912415},{"id":"https://openalex.org/C106516650","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm design","level":2,"score":0.25949999690055847},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.2517000138759613}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/access.2026.3686628","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2026.3686628","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},{"id":"pmh:oai:doaj.org/article:1175b102e011492a98f3be63e69be774","is_oa":true,"landing_page_url":"https://doaj.org/article/1175b102e011492a98f3be63e69be774","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":"IEEE Access, Vol 14, Pp 68271-68280 (2026)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.1109/access.2026.3686628","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2026.3686628","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G3633360255","display_name":null,"funder_award_id":"JPNP20017","funder_id":"https://openalex.org/F4320321034","funder_display_name":"New Energy and Industrial Technology Development Organization"}],"funders":[{"id":"https://openalex.org/F4320321034","display_name":"New Energy and Industrial Technology Development Organization","ror":"https://ror.org/0055k7a87"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Achieving":[0],"fast":[1],"and":[2,17,72,109,141],"accurate":[3],"deep":[4,80],"neural":[5],"network":[6],"(DNN)":[7],"inference":[8,95,128,147],"on":[9,87,96],"edge":[10,97],"devices":[11],"requires":[12],"both":[13],"aggressive":[14],"model":[15],"compression":[16],"efficient":[18],"execution":[19],"pipelines.":[20],"This":[21],"work":[22],"introduces":[23],"a":[24,33,120,126],"sparsity\u2011aware":[25],"compilation":[26],"framework":[27],"that":[28],"integrates":[29],"unstructured":[30,137],"pruning":[31,40],"with":[32,46,107,125],"hybrid":[34],"SparseRT/TensorRT":[35],"optimization":[36],"flow.":[37],"Using":[38],"SynFlow\u2011based":[39],"at":[41],"initialization,":[42],"we":[43],"produce":[44],"models":[45],"over":[47,101],"99.9%":[48],"sparsity":[49],"while":[50,99],"avoiding":[51],"layer":[52],"collapse.":[53],"The":[54],"resulting":[55],"sparse":[56,75],"networks":[57],"are":[58],"compiled":[59],"into":[60],"TensorRT":[61],"engines":[62],"through":[63],"custom":[64],"SparseRT":[65],"kernels,":[66],"fully":[67],"eliminating":[68],"CPU\u2013GPU":[69],"data\u2011transfer":[70],"overhead":[71],"enabling":[73],"end\u2011to\u2011end":[74],"execution.":[76],"Applied":[77],"to":[78],"explainable":[79],"one\u2011class":[81],"classification":[82],"(XDOC)":[83],"for":[84,144],"anomaly":[85],"detection":[86],"MVTec":[88],"AD":[89],"dataset,":[90],"our":[91],"method":[92],"achieves":[93],"sub\u20111\u2011ms":[94],"GPUs":[98],"preserving":[100],"90%":[102],"AUC":[103],"accuracy.":[104],"Replacing":[105],"VGG11\u2011BN":[106],"RepVGG":[108],"incorporating":[110],"CutPaste\u2011augmented":[111],"learning":[112],"further":[113],"increases":[114],"robustness":[115],"under":[116],"extreme":[117],"pruning,":[118,138],"achieving":[119],"parameter":[121],"budget":[122],"of":[123,135],"0.4%":[124],"0.98\u2011ms":[127],"time.":[129],"These":[130],"results":[131],"demonstrate":[132],"the":[133],"effectiveness":[134],"combining":[136],"architecture":[139],"selection,":[140],"compiler\u2011level":[142],"integration":[143],"ultra\u2011low\u2011latency":[145],"edge\u2011AI":[146],"in":[148],"post\u20115G":[149],"digital\u2011twin":[150],"applications.":[151]},"counts_by_year":[],"updated_date":"2026-05-15T08:27:34.491423","created_date":"2026-04-23T00:00:00"}