{"id":"https://openalex.org/W4389934542","doi":"https://doi.org/10.1109/tcsii.2023.3344681","title":"Hardware Accelerator Design for Sparse DNN Inference and Training: A Tutorial","display_name":"Hardware Accelerator Design for Sparse DNN Inference and Training: A Tutorial","publication_year":2023,"publication_date":"2023-12-19","ids":{"openalex":"https://openalex.org/W4389934542","doi":"https://doi.org/10.1109/tcsii.2023.3344681"},"language":"en","primary_location":{"id":"doi:10.1109/tcsii.2023.3344681","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcsii.2023.3344681","pdf_url":null,"source":{"id":"https://openalex.org/S93916849","display_name":"IEEE Transactions on Circuits & Systems II Express Briefs","issn_l":"1549-7747","issn":["1549-7747","1558-3791"],"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 Circuits and Systems II: Express Briefs","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/A5072609627","display_name":"Wendong Mao","orcid":"https://orcid.org/0000-0001-7345-7974"},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wendong Mao","raw_affiliation_strings":["School of Integrated Circuits, Sun Yat-sen University Shenzhen Campus, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Sun Yat-sen University Shenzhen Campus, Shenzhen, China","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100785858","display_name":"Meiqi Wang","orcid":"https://orcid.org/0000-0001-9553-3640"},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Meiqi Wang","raw_affiliation_strings":["School of Integrated Circuits, Sun Yat-sen University Shenzhen Campus, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Sun Yat-sen University Shenzhen Campus, Shenzhen, China","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032640214","display_name":"Xiaoru Xie","orcid":"https://orcid.org/0000-0001-7307-2855"},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaoru Xie","raw_affiliation_strings":["School of Electronic Science and Engineering, Nanjing University, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0001-7307-2855","affiliations":[{"raw_affiliation_string":"School of Electronic Science and Engineering, Nanjing University, Nanjing, China","institution_ids":["https://openalex.org/I881766915"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091293910","display_name":"Xiao Wu","orcid":"https://orcid.org/0000-0001-8306-6889"},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiao Wu","raw_affiliation_strings":["School of Electronic Science and Engineering, Nanjing University, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0001-8306-6889","affiliations":[{"raw_affiliation_string":"School of Electronic Science and Engineering, Nanjing University, Nanjing, China","institution_ids":["https://openalex.org/I881766915"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100696999","display_name":"Zhongfeng Wang","orcid":"https://orcid.org/0000-0002-7227-4786"},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhongfeng Wang","raw_affiliation_strings":["School of Integrated Circuits, Sun Yat-sen University Shenzhen Campus, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Sun Yat-sen University Shenzhen Campus, Shenzhen, China","institution_ids":["https://openalex.org/I157773358"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.1104,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":{"value":0.81916507,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":98},"biblio":{"volume":"71","issue":"3","first_page":"1708","last_page":"1714"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9988999962806702,"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"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9988999962806702,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9976000189781189,"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":0.9907000064849854,"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/computer-science","display_name":"Computer science","score":0.8075977563858032},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6191785335540771},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5366038680076599},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.520885705947876},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.49583300948143005},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4773295521736145},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.47236114740371704},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.4546246826648712},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.4339505434036255},{"id":"https://openalex.org/keywords/sparse-matrix","display_name":"Sparse matrix","score":0.4303799867630005},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.424262672662735},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4232833981513977},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4179559648036957},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.37664729356765747},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.2066296935081482}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8075977563858032},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6191785335540771},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5366038680076599},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.520885705947876},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.49583300948143005},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4773295521736145},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.47236114740371704},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.4546246826648712},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.4339505434036255},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.4303799867630005},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.424262672662735},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4232833981513977},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4179559648036957},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.37664729356765747},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.2066296935081482},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"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/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcsii.2023.3344681","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcsii.2023.3344681","pdf_url":null,"source":{"id":"https://openalex.org/S93916849","display_name":"IEEE Transactions on Circuits & Systems II Express Briefs","issn_l":"1549-7747","issn":["1549-7747","1558-3791"],"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 Circuits and Systems II: Express Briefs","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.8899999856948853}],"awards":[{"id":"https://openalex.org/G4187776313","display_name":null,"funder_award_id":"2022YFB4400604","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"},{"id":"https://openalex.org/G4923597533","display_name":null,"funder_award_id":"2023A007","funder_id":"https://openalex.org/F4320336569","funder_display_name":"Shenzhen Science and Technology Innovation Program"}],"funders":[{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null},{"id":"https://openalex.org/F4320336569","display_name":"Shenzhen Science and Technology Innovation Program","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":66,"referenced_works":["https://openalex.org/W2095705004","https://openalex.org/W2289252105","https://openalex.org/W2304648132","https://openalex.org/W2585720638","https://openalex.org/W2798838583","https://openalex.org/W2883857569","https://openalex.org/W2896457183","https://openalex.org/W2904902077","https://openalex.org/W2921918777","https://openalex.org/W2940345045","https://openalex.org/W2942898010","https://openalex.org/W2945146780","https://openalex.org/W2945514487","https://openalex.org/W2946124490","https://openalex.org/W2963000224","https://openalex.org/W2963840672","https://openalex.org/W2980200167","https://openalex.org/W2984085861","https://openalex.org/W2984140583","https://openalex.org/W2987129023","https://openalex.org/W3016542674","https://openalex.org/W3017024317","https://openalex.org/W3040533934","https://openalex.org/W3040646053","https://openalex.org/W3047848469","https://openalex.org/W3105802176","https://openalex.org/W3106754126","https://openalex.org/W3114479342","https://openalex.org/W3118781290","https://openalex.org/W3127067080","https://openalex.org/W3157114665","https://openalex.org/W3159727696","https://openalex.org/W3159782774","https://openalex.org/W3164217046","https://openalex.org/W3185702163","https://openalex.org/W3187481008","https://openalex.org/W3187677606","https://openalex.org/W3201428215","https://openalex.org/W3206429150","https://openalex.org/W3206453033","https://openalex.org/W3207145355","https://openalex.org/W3210117238","https://openalex.org/W3213229701","https://openalex.org/W4205884968","https://openalex.org/W4214686755","https://openalex.org/W4226345352","https://openalex.org/W4236868170","https://openalex.org/W4285247940","https://openalex.org/W4286571907","https://openalex.org/W4291653336","https://openalex.org/W4293025165","https://openalex.org/W4296415402","https://openalex.org/W4304140796","https://openalex.org/W4308083753","https://openalex.org/W4310997605","https://openalex.org/W4319777991","https://openalex.org/W4360606209","https://openalex.org/W4360770950","https://openalex.org/W4360831795","https://openalex.org/W4360832001","https://openalex.org/W4367663464","https://openalex.org/W4377864779","https://openalex.org/W4379115862","https://openalex.org/W4385976042","https://openalex.org/W4387595847","https://openalex.org/W4387886174"],"related_works":["https://openalex.org/W2952348651","https://openalex.org/W2375742443","https://openalex.org/W2149381099","https://openalex.org/W4200520489","https://openalex.org/W1483190388","https://openalex.org/W2061536531","https://openalex.org/W193873054","https://openalex.org/W2153990583","https://openalex.org/W4387696812","https://openalex.org/W607296746"],"abstract_inverted_index":{"Deep":[0],"neural":[1,93,97],"networks":[2,94,98,102],"(DNNs)":[3],"are":[4],"widely":[5],"used":[6],"in":[7,74,137],"many":[8],"fields,":[9],"such":[10,90],"as":[11,91],"artificial":[12],"intelligence":[13],"generated":[14],"content":[15],"(AIGC)":[16],"and":[17,34,64,78,104,149,175],"robotics.":[18],"To":[19],"efficiently":[20],"support":[21],"these":[22,40],"tasks,":[23],"the":[24,32,57,71,107,143,157],"model":[25,169],"pruning":[26],"technique":[27],"is":[28],"developed":[29],"to":[30,61,124],"compress":[31],"computational":[33],"memory-intensive":[35],"DNNs.":[36],"However,":[37],"directly":[38],"executing":[39],"sparse":[41,58,76,88,119,146,163],"models":[42],"on":[43],"a":[44,111],"common":[45],"hardware":[46,84,126],"accelerator":[47],"can":[48],"cause":[49],"significant":[50],"under-utilization,":[51],"since":[52],"invalid":[53],"data":[54],"resulting":[55],"from":[56,142],"patterns":[59],"leads":[60],"unnecessary":[62],"computations":[63],"irregular":[65],"memory":[66],"accesses.":[67],"This":[68],"brief":[69],"analyzes":[70],"critical":[72],"issues":[73],"accelerating":[75],"models,":[77],"provides":[79],"an":[80],"overview":[81],"of":[82,114,139,145,160],"typical":[83],"designs":[85],"for":[86,118,156],"various":[87],"DNNs,":[89],"convolutional":[92],"(CNNs),":[95],"recurrent":[96],"(RNNs),":[99],"generative":[100],"adversarial":[101],"(GANs),":[103],"Transformers.":[105],"Following":[106],"overview,":[108],"we":[109,133],"give":[110],"practical":[112],"guideline":[113],"designing":[115],"efficient":[116,158],"accelerators":[117],"DNNs":[120],"with":[121,162,172],"qualitative":[122],"metrics":[123],"evaluate":[125],"overhead":[127],"under":[128],"different":[129],"cases.":[130],"In":[131],"addition,":[132],"highlight":[134],"potential":[135],"opportunities":[136],"terms":[138],"hardware/software/algorithm":[140],"co-optimizations":[141],"perspective":[144],"DNN":[147],"implementation,":[148],"provide":[150],"insights":[151],"into":[152],"recent":[153],"design":[154],"trends":[155],"implementation":[159],"transformers":[161],"attention,":[164],"which":[165],"facilitates":[166],"large":[167],"language":[168],"(LLM)":[170],"deployments":[171],"high":[173],"throughput":[174],"energy":[176],"efficiency.":[177]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":4}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}