{"id":"https://openalex.org/W4318685300","doi":"https://doi.org/10.1145/3566097.3567863","title":"Reusing GEMM Hardware for Efficient Execution of Depthwise Separable Convolution on ASIC-Based DNN Accelerators","display_name":"Reusing GEMM Hardware for Efficient Execution of Depthwise Separable Convolution on ASIC-Based DNN Accelerators","publication_year":2023,"publication_date":"2023-01-16","ids":{"openalex":"https://openalex.org/W4318685300","doi":"https://doi.org/10.1145/3566097.3567863"},"language":"en","primary_location":{"id":"doi:10.1145/3566097.3567863","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3566097.3567863","pdf_url":null,"source":{"id":"https://openalex.org/S4363608968","display_name":"Proceedings of the 28th Asia and South Pacific Design Automation Conference","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":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 28th Asia and South Pacific Design Automation Conference","raw_type":"proceedings-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/A5004590102","display_name":"Susmita Dey Manasi","orcid":"https://orcid.org/0000-0001-9358-6255"},"institutions":[{"id":"https://openalex.org/I4210101327","display_name":"Twin Cities Orthopedics","ror":"https://ror.org/01en4s460","country_code":"US","type":"healthcare","lineage":["https://openalex.org/I4210101327"]},{"id":"https://openalex.org/I130238516","display_name":"University of Minnesota","ror":"https://ror.org/017zqws13","country_code":"US","type":"education","lineage":["https://openalex.org/I130238516"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Susmita Dey Manasi","raw_affiliation_strings":["University of Minnesota Twin Cities"],"affiliations":[{"raw_affiliation_string":"University of Minnesota Twin Cities","institution_ids":["https://openalex.org/I4210101327","https://openalex.org/I130238516"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083178603","display_name":"Suvadeep Banerjee","orcid":"https://orcid.org/0000-0001-5188-1651"},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Suvadeep Banerjee","raw_affiliation_strings":["Intel Labs"],"affiliations":[{"raw_affiliation_string":"Intel Labs","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053009380","display_name":"Abhijit Davare","orcid":null},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Abhijit Davare","raw_affiliation_strings":["Intel Labs"],"affiliations":[{"raw_affiliation_string":"Intel Labs","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109435931","display_name":"Anton A. Sorokin","orcid":null},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Anton A. Sorokin","raw_affiliation_strings":["Intel Labs"],"affiliations":[{"raw_affiliation_string":"Intel Labs","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053468539","display_name":"Steven M. Burns","orcid":"https://orcid.org/0000-0003-0248-5403"},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Steven M. Burns","raw_affiliation_strings":["Intel Labs"],"affiliations":[{"raw_affiliation_string":"Intel Labs","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061222925","display_name":"Desmond A. Kirkpatrick","orcid":"https://orcid.org/0000-0002-6982-1830"},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Desmond A. Kirkpatrick","raw_affiliation_strings":["Intel Labs"],"affiliations":[{"raw_affiliation_string":"Intel Labs","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5068714995","display_name":"Sachin S. Sapatnekar","orcid":"https://orcid.org/0000-0002-5353-2364"},"institutions":[{"id":"https://openalex.org/I4210101327","display_name":"Twin Cities Orthopedics","ror":"https://ror.org/01en4s460","country_code":"US","type":"healthcare","lineage":["https://openalex.org/I4210101327"]},{"id":"https://openalex.org/I130238516","display_name":"University of Minnesota","ror":"https://ror.org/017zqws13","country_code":"US","type":"education","lineage":["https://openalex.org/I130238516"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sachin S. Sapatnekar","raw_affiliation_strings":["University of Minnesota Twin Cities"],"affiliations":[{"raw_affiliation_string":"University of Minnesota Twin Cities","institution_ids":["https://openalex.org/I4210101327","https://openalex.org/I130238516"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5004590102"],"corresponding_institution_ids":["https://openalex.org/I130238516","https://openalex.org/I4210101327"],"apc_list":null,"apc_paid":null,"fwci":0.4965,"has_fulltext":false,"cited_by_count":8,"citation_normalized_percentile":{"value":0.57885906,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"475","last_page":"482"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":1.0,"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":1.0,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.996999979019165,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9959999918937683,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.9091336727142334},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8281092047691345},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6943552494049072},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.6691072583198547},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.6562416553497314},{"id":"https://openalex.org/keywords/application-specific-integrated-circuit","display_name":"Application-specific integrated circuit","score":0.6558297276496887},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"Matrix multiplication","score":0.6054412126541138},{"id":"https://openalex.org/keywords/convolution","display_name":"Convolution (computer science)","score":0.5689056515693665},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.5668113231658936},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.4841199815273285},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.4643462300300598},{"id":"https://openalex.org/keywords/multiplication","display_name":"Multiplication (music)","score":0.4564642012119293},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4150620698928833},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.35740697383880615},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.32806363701820374},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.26331278681755066},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.22195714712142944},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.21706122159957886},{"id":"https://openalex.org/keywords/quantum","display_name":"Quantum","score":0.08331531286239624},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.07373583316802979},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.07360509037971497}],"concepts":[{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.9091336727142334},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8281092047691345},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6943552494049072},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.6691072583198547},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.6562416553497314},{"id":"https://openalex.org/C77390884","wikidata":"https://www.wikidata.org/wiki/Q217302","display_name":"Application-specific integrated circuit","level":2,"score":0.6558297276496887},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.6054412126541138},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.5689056515693665},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.5668113231658936},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.4841199815273285},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.4643462300300598},{"id":"https://openalex.org/C2780595030","wikidata":"https://www.wikidata.org/wiki/Q3860309","display_name":"Multiplication (music)","level":2,"score":0.4564642012119293},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4150620698928833},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.35740697383880615},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.32806363701820374},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.26331278681755066},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.22195714712142944},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.21706122159957886},{"id":"https://openalex.org/C84114770","wikidata":"https://www.wikidata.org/wiki/Q46344","display_name":"Quantum","level":2,"score":0.08331531286239624},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.07373583316802979},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.07360509037971497},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"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/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3566097.3567863","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3566097.3567863","pdf_url":null,"source":{"id":"https://openalex.org/S4363608968","display_name":"Proceedings of the 28th Asia and South Pacific Design Automation Conference","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":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 28th Asia and South Pacific Design Automation Conference","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G4790131527","display_name":null,"funder_award_id":"FA8650-20-2-7009","funder_id":"https://openalex.org/F4320332180","funder_display_name":"Defense Advanced Research Projects Agency"}],"funders":[{"id":"https://openalex.org/F4320332180","display_name":"Defense Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":20,"referenced_works":["https://openalex.org/W2531409750","https://openalex.org/W2587013842","https://openalex.org/W2606722458","https://openalex.org/W2612445135","https://openalex.org/W2804032941","https://openalex.org/W2887936511","https://openalex.org/W2919512338","https://openalex.org/W2945146780","https://openalex.org/W2946948417","https://openalex.org/W2963114857","https://openalex.org/W2963125010","https://openalex.org/W2963918968","https://openalex.org/W2989331028","https://openalex.org/W2989762710","https://openalex.org/W3036878841","https://openalex.org/W3042386023","https://openalex.org/W3164610920","https://openalex.org/W3207025137","https://openalex.org/W3216953604","https://openalex.org/W4200413004"],"related_works":["https://openalex.org/W3099313426","https://openalex.org/W4287593139","https://openalex.org/W752783541","https://openalex.org/W1506547947","https://openalex.org/W2995605830","https://openalex.org/W4206811032","https://openalex.org/W2086123442","https://openalex.org/W2532502681","https://openalex.org/W2596457687","https://openalex.org/W3212757063"],"abstract_inverted_index":{"Deep":[0],"learning":[1],"(DL)":[2],"accelerators":[3,55],"are":[4,81],"optimized":[5],"for":[6,83,98,108],"standard":[7,29],"convolution.":[8],"However,":[9],"lightweight":[10,38,99],"convolutional":[11],"neural":[12],"networks":[13],"(CNNs)":[14],"use":[15],"depthwise":[16],"convolution":[17,30],"(DwC)":[18],"in":[19,36],"key":[20],"layers,":[21],"and":[22,28,74,78,103,115,121],"the":[23,46,87],"structural":[24],"difference":[25],"between":[26],"DwC":[27,58,91],"leads":[31],"to":[32,59,69],"significant":[33],"performance":[34,96],"bottleneck":[35],"executing":[37],"CNNs":[39],"on":[40],"such":[41],"platforms.":[42],"This":[43,89],"work":[44],"reuses":[45],"fast":[47],"general":[48],"matrix-vector":[49,62],"multiplication":[50],"(GEMM)":[51],"core":[52],"of":[53,86],"DL":[54,113],"by":[56],"mapping":[57],"channel-wise":[60],"parallel":[61],"multiplications.":[63],"An":[64],"analytical":[65],"framework":[66],"is":[67],"developed":[68,82],"guide":[70],"pre-RTL":[71],"hardware":[72,76],"choices,":[73],"new":[75],"modules":[77],"software":[79],"support":[80],"end-to-end":[84],"evaluation":[85],"solution.":[88],"GEMM-based":[90],"execution":[92],"strategy":[93],"offers":[94],"substantial":[95],"gains":[97],"CNNs:":[100],"7\u00d7":[101],"speedup":[102,117,124],"1.8\u00d7":[104],"lower":[105],"off-chip":[106],"communication":[107],"MobileNet-v1":[109],"over":[110,118,125],"a":[111,119,126],"conventional":[112],"accelerator,":[114],"74\u00d7":[116],"CPU,":[120],"even":[122],"1.4\u00d7":[123],"power-hungry":[127],"GPU.":[128]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":4},{"year":2023,"cited_by_count":2}],"updated_date":"2026-03-28T08:17:26.163206","created_date":"2025-10-10T00:00:00"}