{"id":"https://openalex.org/W2900228909","doi":"https://doi.org/10.1145/3297858.3304028","title":"Packing Sparse Convolutional Neural Networks for Efficient Systolic Array Implementations","display_name":"Packing Sparse Convolutional Neural Networks for Efficient Systolic Array Implementations","publication_year":2019,"publication_date":"2019-04-04","ids":{"openalex":"https://openalex.org/W2900228909","doi":"https://doi.org/10.1145/3297858.3304028","mag":"2900228909"},"language":"en","primary_location":{"id":"doi:10.1145/3297858.3304028","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3297858.3304028","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems","raw_type":"proceedings-article"},"type":"conference-paper","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/A5066867900","display_name":"H. T. Kung","orcid":"https://orcid.org/0000-0002-3348-3788"},"institutions":[{"id":"https://openalex.org/I2801851002","display_name":"Harvard University Press","ror":"https://ror.org/006v7bf86","country_code":"US","type":"other","lineage":["https://openalex.org/I136199984","https://openalex.org/I2801851002"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"H.T. Kung","raw_affiliation_strings":["Harvard University, Cambridge, MA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harvard University, Cambridge, MA, USA","institution_ids":["https://openalex.org/I2801851002"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021927383","display_name":"Bradley McDanel","orcid":"https://orcid.org/0000-0001-6684-8918"},"institutions":[{"id":"https://openalex.org/I2801851002","display_name":"Harvard University Press","ror":"https://ror.org/006v7bf86","country_code":"US","type":"other","lineage":["https://openalex.org/I136199984","https://openalex.org/I2801851002"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bradley McDanel","raw_affiliation_strings":["Harvard University, Cambridge, MA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harvard University, Cambridge, MA, USA","institution_ids":["https://openalex.org/I2801851002"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5013517247","display_name":"Sai Qian Zhang","orcid":"https://orcid.org/0000-0002-4815-9235"},"institutions":[{"id":"https://openalex.org/I2801851002","display_name":"Harvard University Press","ror":"https://ror.org/006v7bf86","country_code":"US","type":"other","lineage":["https://openalex.org/I136199984","https://openalex.org/I2801851002"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sai Qian Zhang","raw_affiliation_strings":["Harvard University, Cambridge, MA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harvard University, Cambridge, MA, USA","institution_ids":["https://openalex.org/I2801851002"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I2801851002"],"apc_list":null,"apc_paid":null,"fwci":9.582,"has_fulltext":false,"cited_by_count":149,"citation_normalized_percentile":{"value":0.98143997,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"821","last_page":"834"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998000264167786,"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":0.9998000264167786,"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9991999864578247,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.9976000189781189,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"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/systolic-array","display_name":"Systolic array","score":0.7309219837188721},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6676703095436096},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.6106813549995422},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.498244047164917},{"id":"https://openalex.org/keywords/application-specific-integrated-circuit","display_name":"Application-specific integrated circuit","score":0.47659987211227417},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.4753686785697937},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.45965972542762756},{"id":"https://openalex.org/keywords/filter","display_name":"Filter (signal processing)","score":0.4479500651359558},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.42353999614715576},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.41390347480773926},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3987198770046234},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.36496034264564514},{"id":"https://openalex.org/keywords/very-large-scale-integration","display_name":"Very-large-scale integration","score":0.26388272643089294},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.25048714876174927},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.18572211265563965},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.11264345049858093},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.09954151511192322},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.09579572081565857}],"concepts":[{"id":"https://openalex.org/C150741067","wikidata":"https://www.wikidata.org/wiki/Q2377218","display_name":"Systolic array","level":3,"score":0.7309219837188721},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6676703095436096},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.6106813549995422},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.498244047164917},{"id":"https://openalex.org/C77390884","wikidata":"https://www.wikidata.org/wiki/Q217302","display_name":"Application-specific integrated circuit","level":2,"score":0.47659987211227417},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4753686785697937},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.45965972542762756},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.4479500651359558},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.42353999614715576},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.41390347480773926},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3987198770046234},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.36496034264564514},{"id":"https://openalex.org/C14580979","wikidata":"https://www.wikidata.org/wiki/Q876049","display_name":"Very-large-scale integration","level":2,"score":0.26388272643089294},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.25048714876174927},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.18572211265563965},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.11264345049858093},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.09954151511192322},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.09579572081565857},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3297858.3304028","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3297858.3304028","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.8999999761581421,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":50,"referenced_works":["https://openalex.org/W575981294","https://openalex.org/W1509443229","https://openalex.org/W1604973310","https://openalex.org/W1686810756","https://openalex.org/W2017369466","https://openalex.org/W2067523571","https://openalex.org/W2096645269","https://openalex.org/W2108598243","https://openalex.org/W2112796928","https://openalex.org/W2121458485","https://openalex.org/W2152839228","https://openalex.org/W2194775991","https://openalex.org/W2198190323","https://openalex.org/W2276486856","https://openalex.org/W2286365479","https://openalex.org/W2289252105","https://openalex.org/W2319920447","https://openalex.org/W2337344472","https://openalex.org/W2466675884","https://openalex.org/W2513554817","https://openalex.org/W2518108298","https://openalex.org/W2525740295","https://openalex.org/W2565851976","https://openalex.org/W2585560244","https://openalex.org/W2585720638","https://openalex.org/W2606722458","https://openalex.org/W2612445135","https://openalex.org/W2613989746","https://openalex.org/W2618530766","https://openalex.org/W2627042741","https://openalex.org/W2736290512","https://openalex.org/W2751366252","https://openalex.org/W2765927961","https://openalex.org/W2767785892","https://openalex.org/W2786618695","https://openalex.org/W2887782013","https://openalex.org/W2899771611","https://openalex.org/W2899915146","https://openalex.org/W2952434909","https://openalex.org/W2962851801","https://openalex.org/W2963000224","https://openalex.org/W2963226883","https://openalex.org/W2963363373","https://openalex.org/W2963993763","https://openalex.org/W2964174152","https://openalex.org/W2964233199","https://openalex.org/W2964299589","https://openalex.org/W3104393472","https://openalex.org/W4212788319","https://openalex.org/W4245199738"],"related_works":["https://openalex.org/W2524802307","https://openalex.org/W2295680811","https://openalex.org/W2359847778","https://openalex.org/W4383501818","https://openalex.org/W3129430558","https://openalex.org/W3140770201","https://openalex.org/W2997828269","https://openalex.org/W2466675884","https://openalex.org/W2951390974","https://openalex.org/W4321192172"],"abstract_inverted_index":{"This":[0],"paper":[1],"describes":[2],"a":[3,13,28,33],"novel":[4],"approach":[5,162],"of":[6,27,45,60,100,122,140,158],"packing":[7],"sparse":[8,25],"convolutional":[9,29],"neural":[10],"networks":[11],"into":[12,32],"denser":[14],"format":[15],"for":[16,72,104,146],"efficient":[17,119],"implementations":[18,121],"using":[19],"systolic":[20,40,47],"arrays.":[21],"By":[22],"combining":[23,70,161],"multiple":[24],"columns":[26],"filter":[30,67,76],"matrix":[31],"single":[34],"dense":[35],"column":[36,160],"stored":[37],"in":[38,63,127],"the":[39,42,46,57,64,79,82,98,132,141,155],"array,":[41],"utilization":[43,107],"efficiency":[44,108,171],"array":[48],"can":[49,134],"be":[50,135],"substantially":[51],"increased":[52,58],"(e.g.,":[53,144],"8x)":[54],"due":[55],"to":[56,92],"density":[59],"nonzero":[61],"weights":[62,77,89],"resulting":[65],"packed":[66],"matrix.":[68],"In":[69],"columns,":[71],"each":[73],"row,":[74],"all":[75],"but":[78],"one":[80],"with":[81,112,137],"largest":[83],"magnitude":[84],"are":[85,90],"pruned.":[86],"The":[87],"remaining":[88],"retrained":[91],"preserve":[93],"high":[94,106],"accuracy.":[95],"We":[96,124,148],"study":[97],"effectiveness":[99],"this":[101],"joint":[102],"optimization":[103],"both":[105],"and":[109,114,151,173],"classification":[110],"accuracy":[111],"ASIC":[113],"FPGA":[115],"designs":[116],"based":[117],"on":[118,154],"bit-serial":[120],"multiplier-accumulators.":[123],"demonstrate":[125],"that":[126],"mitigating":[128],"data":[129],"privacy":[130],"concerns":[131],"retraining":[133],"accomplished":[136],"only":[138],"fractions":[139],"original":[142],"dataset":[143],"10%":[145],"CIFAR-10).":[147],"present":[149],"analysis":[150],"empirical":[152],"evidence":[153],"superior":[156],"performance":[157],"our":[159],"against":[163],"prior":[164],"arts":[165],"under":[166],"metrics":[167],"such":[168],"as":[169],"energy":[170],"(3x)":[172],"inference":[174],"latency":[175],"(12x).":[176]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":14},{"year":2024,"cited_by_count":11},{"year":2023,"cited_by_count":14},{"year":2022,"cited_by_count":29},{"year":2021,"cited_by_count":32},{"year":2020,"cited_by_count":32},{"year":2019,"cited_by_count":12}],"updated_date":"2026-08-21T09:56:20.448147","created_date":"2025-10-10T00:00:00"}