{"id":"https://openalex.org/W4393139774","doi":"https://doi.org/10.1109/tvlsi.2024.3375793","title":"An Energy Efficient Soft SIMD Microarchitecture and Its Application on Quantized CNNs","display_name":"An Energy Efficient Soft SIMD Microarchitecture and Its Application on Quantized CNNs","publication_year":2024,"publication_date":"2024-03-25","ids":{"openalex":"https://openalex.org/W4393139774","doi":"https://doi.org/10.1109/tvlsi.2024.3375793"},"language":"en","primary_location":{"id":"doi:10.1109/tvlsi.2024.3375793","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tvlsi.2024.3375793","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"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 Very Large Scale Integration (VLSI) Systems","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://lirias.kuleuven.be/retrieve/457fd8c4-83d4-4b7a-9f05-d181ec08622e","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5106941408","display_name":"Pengbo Yu","orcid":"https://orcid.org/0009-0004-8221-8608"},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":true,"raw_author_name":"Pengbo Yu","raw_affiliation_strings":["Embedded Systems Laboratory, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland"],"affiliations":[{"raw_affiliation_string":"Embedded Systems Laboratory, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland","institution_ids":["https://openalex.org/I5124864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082759711","display_name":"Flavio Ponzina","orcid":"https://orcid.org/0000-0002-9662-498X"},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]},{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH","US"],"is_corresponding":false,"raw_author_name":"Flavio Ponzina","raw_affiliation_strings":["Embedded Systems Laboratory, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland","University of California at San Diego, La Jolla, CA, USA"],"affiliations":[{"raw_affiliation_string":"Embedded Systems Laboratory, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland","institution_ids":["https://openalex.org/I5124864"]},{"raw_affiliation_string":"University of California at San Diego, La Jolla, CA, USA","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090359025","display_name":"Alexandre Levisse","orcid":"https://orcid.org/0000-0002-8984-9793"},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Alexandre Levisse","raw_affiliation_strings":["Embedded Systems Laboratory, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland"],"affiliations":[{"raw_affiliation_string":"Embedded Systems Laboratory, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland","institution_ids":["https://openalex.org/I5124864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101885160","display_name":"Mohit Gupta","orcid":"https://orcid.org/0000-0002-1924-1264"},"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":"Mohit Gupta","raw_affiliation_strings":["Interuniversity Microelectronics Centre (IMEC), Leuven, Belgium"],"affiliations":[{"raw_affiliation_string":"Interuniversity Microelectronics Centre (IMEC), Leuven, Belgium","institution_ids":["https://openalex.org/I4210114974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060076901","display_name":"Dwaipayan Biswas","orcid":"https://orcid.org/0000-0002-1087-3433"},"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":"Dwaipayan Biswas","raw_affiliation_strings":["Interuniversity Microelectronics Centre (IMEC), Leuven, Belgium"],"affiliations":[{"raw_affiliation_string":"Interuniversity Microelectronics Centre (IMEC), Leuven, Belgium","institution_ids":["https://openalex.org/I4210114974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5044467469","display_name":"Giovanni Ansaloni","orcid":"https://orcid.org/0000-0002-8940-3775"},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Giovanni Ansaloni","raw_affiliation_strings":["Embedded Systems Laboratory, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland"],"affiliations":[{"raw_affiliation_string":"Embedded Systems Laboratory, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland","institution_ids":["https://openalex.org/I5124864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074236306","display_name":"David Atienza","orcid":"https://orcid.org/0000-0001-9536-4947"},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"David Atienza","raw_affiliation_strings":["Embedded Systems Laboratory, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland"],"affiliations":[{"raw_affiliation_string":"Embedded Systems Laboratory, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland","institution_ids":["https://openalex.org/I5124864"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5033235745","display_name":"Francky Catthoor","orcid":"https://orcid.org/0000-0002-3599-8515"},"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":"Francky Catthoor","raw_affiliation_strings":["Interuniversity Microelectronics Centre (IMEC), Leuven, Belgium"],"affiliations":[{"raw_affiliation_string":"Interuniversity Microelectronics Centre (IMEC), Leuven, Belgium","institution_ids":["https://openalex.org/I4210114974"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5106941408"],"corresponding_institution_ids":["https://openalex.org/I5124864"],"apc_list":null,"apc_paid":null,"fwci":2.0726,"has_fulltext":true,"cited_by_count":6,"citation_normalized_percentile":{"value":0.88081747,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":99},"biblio":{"volume":"32","issue":"6","first_page":"1018","last_page":"1031"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9843000173568726,"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/T10320","display_name":"Neural Networks and Applications","score":0.9843000173568726,"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/T13579","display_name":"Image and Video Stabilization","score":0.9372000098228455,"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/T12676","display_name":"Machine Learning and ELM","score":0.9279000163078308,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/simd","display_name":"SIMD","score":0.8623419404029846},{"id":"https://openalex.org/keywords/microarchitecture","display_name":"Microarchitecture","score":0.7548466920852661},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6911584734916687},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6524778008460999},{"id":"https://openalex.org/keywords/very-large-scale-integration","display_name":"Very-large-scale integration","score":0.5222010016441345},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.4308522939682007},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.313434362411499}],"concepts":[{"id":"https://openalex.org/C150552126","wikidata":"https://www.wikidata.org/wiki/Q339387","display_name":"SIMD","level":2,"score":0.8623419404029846},{"id":"https://openalex.org/C107598950","wikidata":"https://www.wikidata.org/wiki/Q259864","display_name":"Microarchitecture","level":2,"score":0.7548466920852661},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6911584734916687},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6524778008460999},{"id":"https://openalex.org/C14580979","wikidata":"https://www.wikidata.org/wiki/Q876049","display_name":"Very-large-scale integration","level":2,"score":0.5222010016441345},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.4308522939682007},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.313434362411499}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/tvlsi.2024.3375793","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tvlsi.2024.3375793","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"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 Very Large Scale Integration (VLSI) Systems","raw_type":"journal-article"},{"id":"pmh:oai:lirias2repo.kuleuven.be:20.500.12942/742413","is_oa":true,"landing_page_url":"https://lirias.kuleuven.be/handle/20.500.12942/742413","pdf_url":"https://lirias.kuleuven.be/retrieve/457fd8c4-83d4-4b7a-9f05-d181ec08622e","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":"Ieee Transactions On Very Large Scale Integration (Vlsi) Systems, vol. 32 (6), (1018-1031)","raw_type":"info:eu-repo/semantics/publishedVersion"},{"id":"pmh:oai:infoscience.epfl.ch:309423","is_oa":true,"landing_page_url":"https://infoscience.epfl.ch/handle/20.500.14299/205898","pdf_url":null,"source":{"id":"https://openalex.org/S4306400487","display_name":"Infoscience (Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"research article"}],"best_oa_location":{"id":"pmh:oai:lirias2repo.kuleuven.be:20.500.12942/742413","is_oa":true,"landing_page_url":"https://lirias.kuleuven.be/handle/20.500.12942/742413","pdf_url":"https://lirias.kuleuven.be/retrieve/457fd8c4-83d4-4b7a-9f05-d181ec08622e","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":"Ieee Transactions On Very Large Scale Integration (Vlsi) Systems, vol. 32 (6), (1018-1031)","raw_type":"info:eu-repo/semantics/publishedVersion"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.8500000238418579,"display_name":"Affordable and clean energy"}],"awards":[{"id":"https://openalex.org/G1185802029","display_name":null,"funder_award_id":"101016776","funder_id":"https://openalex.org/F4320332999","funder_display_name":"Horizon 2020 Framework Programme"}],"funders":[{"id":"https://openalex.org/F4320332999","display_name":"Horizon 2020 Framework Programme","ror":"https://ror.org/00k4n6c32"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4393139774.pdf","grobid_xml":"https://content.openalex.org/works/W4393139774.grobid-xml"},"referenced_works_count":55,"referenced_works":["https://openalex.org/W587823549","https://openalex.org/W634019579","https://openalex.org/W1542024845","https://openalex.org/W1686810756","https://openalex.org/W1805306517","https://openalex.org/W1968221662","https://openalex.org/W2042433781","https://openalex.org/W2042984424","https://openalex.org/W2109897938","https://openalex.org/W2111119192","https://openalex.org/W2112796928","https://openalex.org/W2115778435","https://openalex.org/W2131256658","https://openalex.org/W2143266288","https://openalex.org/W2169721795","https://openalex.org/W2184551777","https://openalex.org/W2194775991","https://openalex.org/W2505041992","https://openalex.org/W2521617467","https://openalex.org/W2536166459","https://openalex.org/W2549139847","https://openalex.org/W2886014761","https://openalex.org/W2913497278","https://openalex.org/W2919636550","https://openalex.org/W2964228333","https://openalex.org/W2966086598","https://openalex.org/W3038347349","https://openalex.org/W3118608800","https://openalex.org/W3123961336","https://openalex.org/W3128854134","https://openalex.org/W3137147200","https://openalex.org/W3144620029","https://openalex.org/W3161051256","https://openalex.org/W3175548485","https://openalex.org/W3177366646","https://openalex.org/W3184863957","https://openalex.org/W4247198796","https://openalex.org/W4287118909","https://openalex.org/W4293025135","https://openalex.org/W4297775537","https://openalex.org/W4317795337","https://openalex.org/W4362510878","https://openalex.org/W4386568763","https://openalex.org/W4386742057","https://openalex.org/W6634030669","https://openalex.org/W6637373629","https://openalex.org/W6684191040","https://openalex.org/W6734062232","https://openalex.org/W6737664043","https://openalex.org/W6753645039","https://openalex.org/W6787972765","https://openalex.org/W6790722212","https://openalex.org/W6791940793","https://openalex.org/W6796815506","https://openalex.org/W6910653704"],"related_works":["https://openalex.org/W2291920536","https://openalex.org/W2162726111","https://openalex.org/W2022397046","https://openalex.org/W1603583590","https://openalex.org/W338671845","https://openalex.org/W2286348849","https://openalex.org/W2018511057","https://openalex.org/W2478126787","https://openalex.org/W1587248296","https://openalex.org/W2063770303"],"abstract_inverted_index":{"The":[0],"ever-increasing":[1],"computational":[2],"complexity":[3],"and":[4,55,141,164,194,214],"energy":[5,163,168,215,242],"consumption":[6,169],"of":[7,21,75],"today\u2019s":[8],"applications,":[9],"such":[10,124],"as":[11,125,191],"machine":[12],"learning":[13],"(ML)":[14],"algorithms,":[15,48],"not":[16],"only":[17,70],"strain":[18],"the":[19,22,33,83,98,188,192],"capabilities":[20],"underlying":[23],"hardware":[24,62],"but":[25],"also":[26,129],"significantly":[27],"restrict":[28],"their":[29],"wide":[30],"deployment":[31],"at":[32,113],"edge.":[34],"Addressing":[35],"these":[36],"challenges,":[37],"novel":[38,89],"architecture":[39],"solutions":[40],"are":[41,158],"required":[42],"by":[43,47],"leveraging":[44],"opportunities":[45],"exposed":[46],"e.g.,":[49],"robustness":[50],"to":[51,139,223,239,245],"small-bitwidth":[52],"operand":[53,76],"quantization":[54],"high":[56],"intrinsic":[57],"data-level":[58],"parallelism.":[59],"However,":[60],"traditional":[61,208],"single":[63],"instruction":[64],"multiple":[65],"data":[66],"(Hard":[67],"SIMD)":[68,102],"architectures":[69],"support":[71],"a":[72,88,161,228],"small":[73],"set":[74],"bitwidths,":[77,235],"limiting":[78],"performance":[79,176],"improvement.":[80],"To":[81],"fill":[82],"gap,":[84],"this":[85],"manuscript":[86],"introduces":[87],"pipelined":[90],"processor":[91],"microarchitecture":[92,116,220],"for":[93,119,175],"arithmetic":[94],"computing":[95],"based":[96],"on":[97,243],"software-defined":[99],"SIMD":[100,108,136,210,230,234],"(Soft":[101],"paradigm":[103],"that":[104,203,231],"can":[105,128],"define":[106],"arbitrary":[107],"modes":[109],"through":[110,173],"control":[111],"instructions":[112],"run-time.":[114],"This":[115],"is":[117,171],"optimized":[118],"parallel":[120],"fine-grained":[121],"fixed-point":[122],"arithmetic,":[123],"shift/add.":[126],"It":[127],"efficiently":[130],"execute":[131,246],"sequential":[132],"shift-add-based":[133],"multiplication":[134],"over":[135],"subwords,":[137],"thanks":[138],"zero-skipping":[140],"canonical":[142],"signed":[143],"digit":[144],"(CSD)":[145],"coding.":[146],"A":[147],"lightweight":[148],"repacking":[149],"unit":[150],"allows":[151],"changing":[152],"subword":[153],"bitwidth":[154],"dynamically.":[155],"These":[156],"features":[157],"implemented":[159],"within":[160],"tight":[162],"area":[165,213,226],"budget.":[166],"An":[167],"model":[170],"established":[172],"post-synthesis":[174],"assessment.":[177],"We":[178],"select":[179],"heterogeneously":[180],"quantized":[181],"(HQ)":[182],"convolutional":[183],"neural":[184],"networks":[185],"(CNNs)":[186],"from":[187],"ML":[189],"domain":[190],"benchmark":[193],"map":[195],"it":[196],"onto":[197],"our":[198,204,219],"microarchitecture.":[199],"Experimental":[200],"results":[201],"showcase":[202],"approach":[205],"dramatically":[206],"outperforms":[207],"Hard":[209,229],"Multiplier-Adder":[211],"regarding":[212],"requirements.":[216],"In":[217],"particular,":[218],"occupies":[221],"up":[222,238],"59.9%":[224],"less":[225,241],"than":[227],"supports":[232],"fewer":[233],"while":[236],"consuming":[237],"50.1%":[240],"average":[244],"HQ":[247],"CNNs.":[248]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":1}],"updated_date":"2026-04-02T15:55:50.835912","created_date":"2025-10-10T00:00:00"}