{"id":"https://openalex.org/W3114479342","doi":"https://doi.org/10.1109/jssc.2020.3043870","title":"SNAP: An Efficient Sparse Neural Acceleration Processor for Unstructured Sparse Deep Neural Network Inference","display_name":"SNAP: An Efficient Sparse Neural Acceleration Processor for Unstructured Sparse Deep Neural Network Inference","publication_year":2020,"publication_date":"2020-12-29","ids":{"openalex":"https://openalex.org/W3114479342","doi":"https://doi.org/10.1109/jssc.2020.3043870","mag":"3114479342"},"language":"en","primary_location":{"id":"doi:10.1109/jssc.2020.3043870","doi":"https://doi.org/10.1109/jssc.2020.3043870","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jssc.2020.3043870","pdf_url":null,"source":{"id":"https://openalex.org/S83637746","display_name":"IEEE Journal of Solid-State Circuits","issn_l":"0018-9200","issn":["0018-9200","1558-173X"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","dongbi-a","jufo-3","norway-2","poland-140","russia-white-list-1"],"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 Journal of Solid-State Circuits","raw_type":"journal-article"},"parent_works":[],"lineage":["https://openalex.org/W3114479342"],"child_works_count":0,"type":"article","is_traditional":true,"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/A5076596358","display_name":"Zhang Jie-Fang\ue00b","orcid":"https://orcid.org/0000-0002-6609-4383","observed_orcids":["https://orcid.org/0000-0002-6609-4383"]},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jie-Fang Zhang","raw_affiliation_strings":["University of Michigan, Ann Arbor, MI, USA"],"raw_orcid":"https://orcid.org/0000-0002-6609-4383","affiliations":[{"raw_affiliation_string":"University of Michigan, Ann Arbor, MI, USA","institution_ids":["https://openalex.org/I27837315"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007101760","display_name":"Ching\u2010En Lee","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ching-En Lee","raw_affiliation_strings":["University of Michigan, Ann Arbor, MI, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Michigan, Ann Arbor, MI, USA","institution_ids":["https://openalex.org/I27837315"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041944874","display_name":"Chester Liu","orcid":"https://orcid.org/0000-0003-0115-9630","observed_orcids":["https://orcid.org/0000-0003-0115-9630"]},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chester Liu","raw_affiliation_strings":["University of Michigan, Ann Arbor, MI, USA"],"raw_orcid":"https://orcid.org/0000-0003-0115-9630","affiliations":[{"raw_affiliation_string":"University of Michigan, Ann Arbor, MI, USA","institution_ids":["https://openalex.org/I27837315"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008683881","display_name":"Yakun Sophia Shao","orcid":"https://orcid.org/0000-0003-1811-5407","observed_orcids":["https://orcid.org/0000-0003-1811-5407"]},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]},{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yakun Sophia Shao","raw_affiliation_strings":["Nvidia Research, Santa Clara, CA, USA","University of California at Berkeley, Berkeley, CA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nvidia Research, Santa Clara, CA, USA","institution_ids":["https://openalex.org/I4210127875"]},{"raw_affiliation_string":"University of California at Berkeley, Berkeley, CA, USA","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063354509","display_name":"Stephen W. Keckler","orcid":"https://orcid.org/0000-0001-6701-6099","observed_orcids":["https://orcid.org/0000-0001-6701-6099"]},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Stephen W. Keckler","raw_affiliation_strings":["NVIDIA, Austin, TX, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA, Austin, TX, USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5038110244","display_name":"Zhengya Zhang","orcid":"https://orcid.org/0000-0001-5963-9018","observed_orcids":["https://orcid.org/0000-0001-5963-9018"]},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhengya Zhang","raw_affiliation_strings":["University of Michigan, Ann Arbor, MI, USA"],"raw_orcid":"https://orcid.org/0000-0001-5963-9018","affiliations":[{"raw_affiliation_string":"University of Michigan, Ann Arbor, MI, USA","institution_ids":["https://openalex.org/I27837315"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":{"value":2045,"currency":"USD","value_usd":2045},"apc_paid":null,"fwci":3.4586,"has_fulltext":false,"cited_by_count":111,"citation_normalized_percentile":{"value":0.93824628,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":100},"biblio":{"volume":"56","issue":"2","first_page":"636","last_page":"647"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.923799991607666,"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":0.923799991607666,"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":0.03709999844431877,"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/T10363","display_name":"Low-power high-performance VLSI design","score":0.01600000075995922,"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/unstructured-sparsity","display_name":"unstructured sparsity","score":0.9710000157356262},{"id":"https://openalex.org/keywords/deep-neural-network-inference","display_name":"deep neural network inference","score":0.722000002861023},{"id":"https://openalex.org/keywords/neural-network-pruning","display_name":"neural network pruning","score":0.6970000267028809},{"id":"https://openalex.org/keywords/associative-search","display_name":"associative search","score":0.8130000233650208},{"id":"https://openalex.org/keywords/convolutional-layers","display_name":"convolutional layers","score":0.5149999856948853},{"id":"https://openalex.org/keywords/pointwise-convolution","display_name":"pointwise convolution","score":0.7940000295639038},{"id":"https://openalex.org/keywords/fully-connected-layers","display_name":"fully connected layers","score":0.9300000071525574},{"id":"https://openalex.org/keywords/energy-efficient-hardware","display_name":"energy-efficient hardware","score":0.6330000162124634},{"id":"https://openalex.org/keywords/resnet-50","display_name":"ResNet50","score":0.7160000205039978},{"id":"https://openalex.org/keywords/16-nm-cmos","display_name":"16nm CMOS","score":0.8100000023841858}],"concepts":[{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.8008165955543518},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7458822131156921},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6362330317497253},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5079441666603088},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.4315207898616791},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.42758437991142273},{"id":"https://openalex.org/C165005293","wikidata":"https://www.wikidata.org/wiki/Q1074500","display_name":"Chip","level":2,"score":0.4232257008552551},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.42099729180336},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.4133108854293823},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.4108608365058899},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.36735740303993225},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.21434754133224487},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.2112216353416443},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.10587820410728455},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/jssc.2020.3043870","doi":"https://doi.org/10.1109/jssc.2020.3043870","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jssc.2020.3043870","pdf_url":null,"source":{"id":"https://openalex.org/S83637746","display_name":"IEEE Journal of Solid-State Circuits","issn_l":"0018-9200","issn":["0018-9200","1558-173X"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","dongbi-a","jufo-3","norway-2","poland-140","russia-white-list-1"],"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 Journal of Solid-State Circuits","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320315934","display_name":"Toyota Research Institute","ror":"https://ror.org/04fpkc108"},{"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":31,"referenced_works":["https://openalex.org/W1483870316","https://openalex.org/W1686810756","https://openalex.org/W1903029394","https://openalex.org/W2102605133","https://openalex.org/W2108598243","https://openalex.org/W2119144962","https://openalex.org/W2163605009","https://openalex.org/W2183341477","https://openalex.org/W2194775991","https://openalex.org/W2285660444","https://openalex.org/W2289252105","https://openalex.org/W2442974303","https://openalex.org/W2516141709","https://openalex.org/W2541839172","https://openalex.org/W2565851976","https://openalex.org/W2593564159","https://openalex.org/W2606722458","https://openalex.org/W2625457103","https://openalex.org/W2798170643","https://openalex.org/W2895531329","https://openalex.org/W2899481200","https://openalex.org/W2919115771","https://openalex.org/W2945146780","https://openalex.org/W2963674932","https://openalex.org/W2964299589","https://openalex.org/W2964988320","https://openalex.org/W2987129023","https://openalex.org/W3023154643","https://openalex.org/W4236868170","https://openalex.org/W4240168186","https://openalex.org/W6734592959"],"related_works":["https://openalex.org/W2964988320","https://openalex.org/W2962816593","https://openalex.org/W4411485108","https://openalex.org/W4410893427","https://openalex.org/W4405022794","https://openalex.org/W4406931554","https://openalex.org/W3105802176","https://openalex.org/W2886352393","https://openalex.org/W4380303214","https://openalex.org/W4286841766"],"abstract_inverted_index":{"Recent":[0],"developments":[1],"in":[2,62,143,164],"deep":[3,13],"neural":[4,54],"network":[5],"(DNN)":[6],"pruning":[7],"introduces":[8],"data":[9,86],"sparsity":[10,36,61],"to":[11,16,33,43,58,69,93,114,146,175],"enable":[12],"learning":[14],"applications":[15],"run":[17],"more":[18],"efficiently":[19],"on":[20,202],"resourceand":[21],"energy-constrained":[22],"hardware":[23,31],"platforms.":[24],"However,":[25],"these":[26],"sparse":[27,53,82],"models":[28],"require":[29],"specialized":[30],"structures":[32],"exploit":[34,59],"the":[35,44,52,119,124,207],"for":[37,192],"storage,":[38],"latency,":[39],"and":[40,74,84,105,122,154,189,198,220],"efficiency":[41,180],"improvements":[42],"full":[45],"extent.":[46],"In":[47],"this":[48],"work,":[49],"we":[50],"present":[51],"acceleration":[55],"processor":[56],"(SNAP)":[57],"unstructured":[60],"DNNs.":[63],"SNAP":[64,92,100,160],"uses":[65,106],"parallel":[66],"associative":[67,89],"search":[68,90],"discover":[70],"valid":[71],"weight":[72,197],"(W)":[73],"input":[75],"activation":[76,199],"(IA)":[77],"pairs":[78],"from":[79],"compressed,":[80],"unstructured,":[81],"W":[83],"IA":[85],"arrays.":[87],"The":[88,169],"allows":[91],"maintain":[94],"a":[95,102,107,165,177,203,211],"75%":[96],"average":[97],"compute":[98],"utilization.":[99],"follows":[101],"channel-first":[103],"dataflow":[104,113,139],"two-level":[108],"partial":[109],"sum":[110],"(psum)":[111],"reduction":[112,138],"eliminate":[115],"access":[116],"contention":[117],"at":[118,186,217],"output":[120],"buffer":[121],"cut":[123],"psum":[125,137],"writeback":[126],"traffic":[127],"by":[128],"22\u00d7":[129],"compared":[130],"with":[131,195],"state-of-the-art":[132],"DNN":[133],"accelerator":[134],"designs.":[135],"SNAP's":[136],"can":[140],"be":[141],"configured":[142],"two":[144],"modes":[145],"support":[147],"general":[148],"convolution":[149],"(CONV)":[150],"layers,":[151],"pointwise":[152],"CONV,":[153],"fully":[155],"connected":[156],"layers.":[157],"A":[158],"prototype":[159],"chip":[161,172,209],"is":[162,173],"implemented":[163],"16-nm":[166],"CMOS":[167],"technology.":[168],"2.3-mm2":[170],"test":[171,208],"measured":[174],"achieve":[176],"peak":[178,212],"effectual":[179],"of":[181,214],"21.55":[182],"TOPS/W":[183],"(16":[184],"b)":[185],"0.55":[187],"V":[188,219],"260":[190],"MHz":[191],"CONV":[193],"layers":[194],"10%":[196],"densities.":[200],"Operating":[201],"pruned":[204],"ResNet-50":[205],"network,":[206],"achieves":[210],"throughput":[213],"90.98":[215],"frames/s":[216],"0.80":[218],"480":[221],"MHz,":[222],"dissipating":[223],"348":[224],"mW.":[225]},"counts_by_year":[{"year":2026,"cited_by_count":17},{"year":2025,"cited_by_count":32},{"year":2024,"cited_by_count":21},{"year":2023,"cited_by_count":25},{"year":2022,"cited_by_count":12},{"year":2021,"cited_by_count":4}],"updated_date":"2026-11-14T11:04:58.165337","created_date":"2021-01-05T00:00:00"}