{"id":"https://openalex.org/W2964504576","doi":"https://doi.org/10.1109/aicas.2019.8771481","title":"Sub-Word Parallel Precision-Scalable MAC Engines for Efficient Embedded DNN Inference","display_name":"Sub-Word Parallel Precision-Scalable MAC Engines for Efficient Embedded DNN Inference","publication_year":2019,"publication_date":"2019-03-01","ids":{"openalex":"https://openalex.org/W2964504576","doi":"https://doi.org/10.1109/aicas.2019.8771481","mag":"2964504576"},"language":"en","primary_location":{"id":"doi:10.1109/aicas.2019.8771481","is_oa":false,"landing_page_url":"https://doi.org/10.1109/aicas.2019.8771481","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2019 IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://lirias.kuleuven.be/handle/123456789/640380","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5017238376","display_name":"Linyan Mei","orcid":"https://orcid.org/0000-0001-8649-3923"},"institutions":[{"id":"https://openalex.org/I99464096","display_name":"KU Leuven","ror":"https://ror.org/05f950310","country_code":"BE","type":"education","lineage":["https://openalex.org/I99464096"]}],"countries":["BE"],"is_corresponding":true,"raw_author_name":"Linyan Mei","raw_affiliation_strings":["Dept. of Electrical Engineering, MICAS, KU Leuven, Belgium"],"affiliations":[{"raw_affiliation_string":"Dept. of Electrical Engineering, MICAS, KU Leuven, Belgium","institution_ids":["https://openalex.org/I99464096"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014773761","display_name":"Mohit Dandekar","orcid":"https://orcid.org/0000-0002-8713-0945"},"institutions":[{"id":"https://openalex.org/I99464096","display_name":"KU Leuven","ror":"https://ror.org/05f950310","country_code":"BE","type":"education","lineage":["https://openalex.org/I99464096"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Mohit Dandekar","raw_affiliation_strings":["Dept. of Electrical Engineering, MICAS, KU Leuven, Belgium"],"affiliations":[{"raw_affiliation_string":"Dept. of Electrical Engineering, MICAS, KU Leuven, Belgium","institution_ids":["https://openalex.org/I99464096"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028512422","display_name":"Dimitrios Rodopoulos","orcid":"https://orcid.org/0000-0002-1408-3658"},"institutions":[{"id":"https://openalex.org/I4210114974","display_name":"IMEC","ror":"https://ror.org/02kcbn207","country_code":"BE","type":"nonprofit","lineage":["https://openalex.org/I4210114974"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Dimitrios Rodopoulos","raw_affiliation_strings":["imec, Belgium"],"affiliations":[{"raw_affiliation_string":"imec, Belgium","institution_ids":["https://openalex.org/I4210114974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111593170","display_name":"Jeremy Constantin","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114974","display_name":"IMEC","ror":"https://ror.org/02kcbn207","country_code":"BE","type":"nonprofit","lineage":["https://openalex.org/I4210114974"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Jeremy Constantin","raw_affiliation_strings":["imec, Belgium"],"affiliations":[{"raw_affiliation_string":"imec, Belgium","institution_ids":["https://openalex.org/I4210114974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082798974","display_name":"Peter Debacker","orcid":"https://orcid.org/0000-0003-3825-5554"},"institutions":[{"id":"https://openalex.org/I4210114974","display_name":"IMEC","ror":"https://ror.org/02kcbn207","country_code":"BE","type":"nonprofit","lineage":["https://openalex.org/I4210114974"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Peter Debacker","raw_affiliation_strings":["imec, Belgium"],"affiliations":[{"raw_affiliation_string":"imec, Belgium","institution_ids":["https://openalex.org/I4210114974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039430610","display_name":"Rudy Lauwereins","orcid":"https://orcid.org/0000-0002-3861-0168"},"institutions":[{"id":"https://openalex.org/I4210114974","display_name":"IMEC","ror":"https://ror.org/02kcbn207","country_code":"BE","type":"nonprofit","lineage":["https://openalex.org/I4210114974"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Rudy Lauwereins","raw_affiliation_strings":["imec, Belgium"],"affiliations":[{"raw_affiliation_string":"imec, Belgium","institution_ids":["https://openalex.org/I4210114974"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5012150553","display_name":"Marian Verhelst","orcid":"https://orcid.org/0000-0003-3495-9263"},"institutions":[{"id":"https://openalex.org/I99464096","display_name":"KU Leuven","ror":"https://ror.org/05f950310","country_code":"BE","type":"education","lineage":["https://openalex.org/I99464096"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Marian Verhelst","raw_affiliation_strings":["Dept. of Electrical Engineering, MICAS, KU Leuven, Belgium"],"affiliations":[{"raw_affiliation_string":"Dept. of Electrical Engineering, MICAS, KU Leuven, Belgium","institution_ids":["https://openalex.org/I99464096"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5017238376"],"corresponding_institution_ids":["https://openalex.org/I99464096"],"apc_list":null,"apc_paid":null,"fwci":1.7208,"has_fulltext":false,"cited_by_count":40,"citation_normalized_percentile":{"value":0.87932081,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"6","last_page":"10"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9997000098228455,"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":0.9997000098228455,"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":0.9990000128746033,"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":0.9983999729156494,"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/scalability","display_name":"Scalability","score":0.7760860919952393},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7622218728065491},{"id":"https://openalex.org/keywords/multiplier","display_name":"Multiplier (economics)","score":0.6634251475334167},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.5971883535385132},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.539381742477417},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4882690906524658},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4776153266429901},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.475276380777359},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.44292497634887695},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.3894810080528259},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.38928577303886414},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3815631568431854},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.367361843585968},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.15140235424041748},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1466217041015625}],"concepts":[{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7760860919952393},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7622218728065491},{"id":"https://openalex.org/C124584101","wikidata":"https://www.wikidata.org/wiki/Q1053266","display_name":"Multiplier (economics)","level":2,"score":0.6634251475334167},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.5971883535385132},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.539381742477417},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4882690906524658},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4776153266429901},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.475276380777359},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.44292497634887695},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3894810080528259},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.38928577303886414},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3815631568431854},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.367361843585968},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.15140235424041748},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1466217041015625},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","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},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","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/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0},{"id":"https://openalex.org/C139719470","wikidata":"https://www.wikidata.org/wiki/Q39680","display_name":"Macroeconomics","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/aicas.2019.8771481","is_oa":false,"landing_page_url":"https://doi.org/10.1109/aicas.2019.8771481","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2019 IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS)","raw_type":"proceedings-article"},{"id":"pmh:oai:lirias2repo.kuleuven.be:123456789/640380","is_oa":true,"landing_page_url":"https://lirias.kuleuven.be/handle/123456789/640380","pdf_url":null,"source":{"id":"https://openalex.org/S4306401954","display_name":"Lirias (KU Leuven)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I99464096","host_organization_name":"KU Leuven","host_organization_lineage":["https://openalex.org/I99464096"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"2019 IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS), Hsinchu 30060, Taiwan, 18-20 March 2019","raw_type":"info:eu-repo/semantics/publishedVersion"}],"best_oa_location":{"id":"pmh:oai:lirias2repo.kuleuven.be:123456789/640380","is_oa":true,"landing_page_url":"https://lirias.kuleuven.be/handle/123456789/640380","pdf_url":null,"source":{"id":"https://openalex.org/S4306401954","display_name":"Lirias (KU Leuven)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I99464096","host_organization_name":"KU Leuven","host_organization_lineage":["https://openalex.org/I99464096"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"2019 IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS), Hsinchu 30060, Taiwan, 18-20 March 2019","raw_type":"info:eu-repo/semantics/publishedVersion"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.9100000262260437,"display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W2308155233","https://openalex.org/W2337344472","https://openalex.org/W2557321428","https://openalex.org/W2593564159","https://openalex.org/W2594492285","https://openalex.org/W2604319603","https://openalex.org/W2768923469","https://openalex.org/W2769502706","https://openalex.org/W2774659647","https://openalex.org/W2794141774","https://openalex.org/W2800022361","https://openalex.org/W2890770683","https://openalex.org/W2892163332","https://openalex.org/W2962717701","https://openalex.org/W2963093735","https://openalex.org/W2963367920","https://openalex.org/W2963451457","https://openalex.org/W2963921921","https://openalex.org/W2965976751","https://openalex.org/W4294565153","https://openalex.org/W6703414193","https://openalex.org/W6730230071","https://openalex.org/W6746233725","https://openalex.org/W6746307771","https://openalex.org/W6746658245"],"related_works":["https://openalex.org/W2136583354","https://openalex.org/W2111238207","https://openalex.org/W2760721665","https://openalex.org/W330130819","https://openalex.org/W2288610023","https://openalex.org/W2112044895","https://openalex.org/W3121416282","https://openalex.org/W2281389338","https://openalex.org/W2037453743","https://openalex.org/W3195229121"],"abstract_inverted_index":{"To":[0],"enable":[1],"energy-efficient":[2,45],"embedded":[3],"execution":[4],"of":[5,13,41,74,87,103,134,154],"Deep":[6],"Neural":[7],"Networks":[8],"(DNNs),":[9],"the":[10,37,50,71,88,101,116,131,140,144,151,171,189,195],"critical":[11],"sections":[12],"these":[14],"workloads,":[15],"their":[16],"multiply-accumulate":[17],"(MAC)":[18],"operations,":[19],"need":[20],"to":[21,48,93,163,200],"be":[22,114],"carefully":[23],"optimized.":[24],"The":[25,167,185],"SotA":[26,56,91],"pursues":[27],"this":[28,123],"through":[29],"run-time":[30],"precision-scalable":[31],"MAC":[32,53,155],"operators,":[33],"which":[34],"can":[35],"support":[36,94],"varying":[38],"precision":[39,52,63,106,137],"needs":[40],"DNNs":[42],"in":[43,176],"an":[44,126],"way.":[46],"Yet,":[47],"implement":[49,164],"adaptable":[51],"operation,":[54],"most":[55,89],"solutions":[57],"rely":[58],"on":[59,150,182,202,208],"separately":[60],"optimized":[61],"low":[62,136],"multipliers":[64],"and":[65,79,206],"a":[66,75,104],"precision-variable":[67],"accumulation":[68,133],"scheme,":[69,129],"with":[70],"possible":[72],"disadvantages":[73],"high":[76,105],"control":[77],"complexity":[78],"degraded":[80],"throughput.":[81],"This":[82,98],"paper,":[83],"first":[84],"optimizes":[85],"one":[86],"effective":[90],"techniques":[92],"fully-connected":[95],"DNN":[96,165],"layers.":[97],"mode,":[99],"exploiting":[100],"transformation":[102],"multiplier":[107,141],"into":[108],"independent":[109],"parallel":[110],"low-precision":[111,128],"multipliers,":[112],"will":[113],"called":[115,143],"Sum":[117,145],"Separate":[118],"(SS)":[119],"mode.":[120,148],"In":[121],"addition,":[122],"work":[124],"suggests":[125],"alternative":[127],"i.e.":[130],"implicit":[132],"multiple":[135],"products":[138],"within":[139],"itself,":[142],"Together":[146],"(ST)":[147],"Based":[149],"two":[152,168],"types":[153],"arrangements":[156],"explored,":[157],"corresponding":[158],"architectures":[159],"have":[160],"been":[161],"proposed":[162,190],"processing.":[166],"architectures,":[169],"yielding":[170],"same":[172],"throughput,":[173],"are":[174],"compared":[175],"different":[177],"working":[178],"precisions":[179],"(2/4/8/16-bit),":[180],"based":[181,192],"Post-Synthesis":[183],"simulation.":[184],"result":[186],"shows":[187],"that":[188],"ST-Mode":[191],"architecture":[193],"outperforms":[194],"earlier":[196],"SS-Mode":[197],"by":[198],"up":[199],"\u00d71.6":[201],"Energy":[203],"Efficiency":[204,210],"(TOPS/W)":[205],"\u00d71.5":[207],"Area":[209],"(GOPS/mm":[211],"<sup":[212],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[213],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">2</sup>":[214],").":[215]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":8},{"year":2023,"cited_by_count":11},{"year":2022,"cited_by_count":8},{"year":2021,"cited_by_count":3},{"year":2020,"cited_by_count":4},{"year":2019,"cited_by_count":2}],"updated_date":"2026-02-11T14:41:00.668223","created_date":"2025-10-10T00:00:00"}