{"id":"https://openalex.org/W4386474973","doi":"https://doi.org/10.1109/isvlsi59464.2023.10238679","title":"A 3 TOPS/W RISC-V Parallel Cluster for Inference of Fine-Grain Mixed-Precision Quantized Neural Networks","display_name":"A 3 TOPS/W RISC-V Parallel Cluster for Inference of Fine-Grain Mixed-Precision Quantized Neural Networks","publication_year":2023,"publication_date":"2023-06-20","ids":{"openalex":"https://openalex.org/W4386474973","doi":"https://doi.org/10.1109/isvlsi59464.2023.10238679"},"language":"en","primary_location":{"id":"doi:10.1109/isvlsi59464.2023.10238679","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isvlsi59464.2023.10238679","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://cris.unibo.it/bitstream/11585/953207/5/a%2b3%2btops%2bw%2brisc%2bpost%2bprint%2b.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5054770753","display_name":"Alessandro Nadalini","orcid":"https://orcid.org/0009-0007-3574-7576"},"institutions":[{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Alessandro Nadalini","raw_affiliation_strings":["University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","institution_ids":["https://openalex.org/I9360294"]},{"raw_affiliation_string":"Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007387832","display_name":"Georg Rutishauser","orcid":"https://orcid.org/0000-0001-8875-7611"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Georg Rutishauser","raw_affiliation_strings":["IIS Integrated Systems Laboratory, ETH Zurich,Switzerland","IIS Integrated Systems Laboratory, ETH Zurich, Switzerland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IIS Integrated Systems Laboratory, ETH Zurich,Switzerland","institution_ids":["https://openalex.org/I35440088"]},{"raw_affiliation_string":"IIS Integrated Systems Laboratory, ETH Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032095821","display_name":"Alessio Burrello","orcid":"https://orcid.org/0000-0002-6215-8220"},"institutions":[{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Alessio Burrello","raw_affiliation_strings":["University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","institution_ids":["https://openalex.org/I9360294"]},{"raw_affiliation_string":"Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003115244","display_name":"Nazareno Bruschi","orcid":null},"institutions":[{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Nazareno Bruschi","raw_affiliation_strings":["University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","institution_ids":["https://openalex.org/I9360294"]},{"raw_affiliation_string":"Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052915995","display_name":"Angelo Garofalo","orcid":"https://orcid.org/0000-0002-7495-6895"},"institutions":[{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Angelo Garofalo","raw_affiliation_strings":["University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","institution_ids":["https://openalex.org/I9360294"]},{"raw_affiliation_string":"Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043408422","display_name":"Luca Benini","orcid":"https://orcid.org/0000-0001-8068-3806"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]},{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["CH","IT"],"is_corresponding":false,"raw_author_name":"Luca Benini","raw_affiliation_strings":["University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy","IIS Integrated Systems Laboratory, ETH Zurich, Switzerland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","institution_ids":["https://openalex.org/I9360294"]},{"raw_affiliation_string":"Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]},{"raw_affiliation_string":"IIS Integrated Systems Laboratory, ETH Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038717922","display_name":"Francesco Conti","orcid":"https://orcid.org/0000-0002-7924-933X"},"institutions":[{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Francesco Conti","raw_affiliation_strings":["University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","institution_ids":["https://openalex.org/I9360294"]},{"raw_affiliation_string":"Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5023905268","display_name":"Davide Rossi","orcid":"https://orcid.org/0000-0002-0651-5393"},"institutions":[{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Davide Rossi","raw_affiliation_strings":["University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Bologna,Department of Electrical, Electronic and Information Engineering (DEI),Italy","institution_ids":["https://openalex.org/I9360294"]},{"raw_affiliation_string":"Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.0636,"has_fulltext":true,"cited_by_count":10,"citation_normalized_percentile":{"value":0.79404283,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9995999932289124,"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":0.9995999932289124,"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/T12702","display_name":"Brain Tumor Detection and Classification","score":0.995199978351593,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9944000244140625,"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":0.8334941864013672},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5338199138641357},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.5041357278823853},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.46106797456741333},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.44371742010116577},{"id":"https://openalex.org/keywords/compiler","display_name":"Compiler","score":0.4201151132583618},{"id":"https://openalex.org/keywords/static-random-access-memory","display_name":"Static random-access memory","score":0.4154777526855469},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.3829059898853302},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.3726844787597656},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3675132393836975},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.35160669684410095},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.12248682975769043},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.120364248752594}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8334941864013672},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5338199138641357},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.5041357278823853},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.46106797456741333},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.44371742010116577},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.4201151132583618},{"id":"https://openalex.org/C68043766","wikidata":"https://www.wikidata.org/wiki/Q267416","display_name":"Static random-access memory","level":2,"score":0.4154777526855469},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3829059898853302},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.3726844787597656},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3675132393836975},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.35160669684410095},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.12248682975769043},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.120364248752594},{"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/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/isvlsi59464.2023.10238679","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isvlsi59464.2023.10238679","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)","raw_type":"proceedings-article"},{"id":"pmh:oai:cris.unibo.it:11585/953207","is_oa":true,"landing_page_url":"https://hdl.handle.net/11585/953207","pdf_url":"https://cris.unibo.it/bitstream/11585/953207/5/a%2b3%2btops%2bw%2brisc%2bpost%2bprint%2b.pdf","source":{"id":"https://openalex.org/S4306402579","display_name":"Archivio istituzionale della ricerca (Alma Mater Studiorum Universit\u00e0 di Bologna)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4210117483","host_organization_name":"Istituto di Ematologia di Bologna","host_organization_lineage":["https://openalex.org/I4210117483"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":{"id":"pmh:oai:cris.unibo.it:11585/953207","is_oa":true,"landing_page_url":"https://hdl.handle.net/11585/953207","pdf_url":"https://cris.unibo.it/bitstream/11585/953207/5/a%2b3%2btops%2bw%2brisc%2bpost%2bprint%2b.pdf","source":{"id":"https://openalex.org/S4306402579","display_name":"Archivio istituzionale della ricerca (Alma Mater Studiorum Universit\u00e0 di Bologna)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4210117483","host_organization_name":"Istituto di Ematologia di Bologna","host_organization_lineage":["https://openalex.org/I4210117483"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/conferenceObject"},"sustainable_development_goals":[{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.8899999856948853}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4386474973.pdf"},"referenced_works_count":21,"referenced_works":["https://openalex.org/W2276486856","https://openalex.org/W2524428287","https://openalex.org/W2565125333","https://openalex.org/W2784372305","https://openalex.org/W2794141774","https://openalex.org/W2963255460","https://openalex.org/W2970601456","https://openalex.org/W2982041622","https://openalex.org/W3013186616","https://openalex.org/W3032746325","https://openalex.org/W3047342117","https://openalex.org/W3049586678","https://openalex.org/W3102860086","https://openalex.org/W3137875211","https://openalex.org/W3156145546","https://openalex.org/W3203242999","https://openalex.org/W4220837341","https://openalex.org/W6727208969","https://openalex.org/W6747876305","https://openalex.org/W6762754082","https://openalex.org/W6767298317"],"related_works":["https://openalex.org/W3151633427","https://openalex.org/W3096456556","https://openalex.org/W4240253816","https://openalex.org/W2212894501","https://openalex.org/W2793465010","https://openalex.org/W3024050170","https://openalex.org/W2109451123","https://openalex.org/W4293253840","https://openalex.org/W4378977321","https://openalex.org/W3013241373"],"abstract_inverted_index":{"The":[0],"emerging":[1],"trend":[2],"of":[3,63,95,119,133,176,188],"deploying":[4],"complex":[5],"algorithms,":[6],"such":[7],"as":[8,30],"Deep":[9],"Neural":[10,66],"networks":[11],"(DNNs),":[12],"increasingly":[13],"poses":[14],"strict":[15],"memory":[16,37],"and":[17,39,57,140,152,172],"energy":[18],"efficiency":[19],"requirements":[20],"on":[21,75,155],"Internet-of-Things":[22],"(IoT)":[23],"end-nodes.":[24],"Mixed-precision":[25],"quantization":[26],"has":[27],"been":[28],"proposed":[29],"a":[31,35,54,72,116,126,136,141,160],"technique":[32],"to":[33,90,100,149,167,181,207],"minimize":[34],"DNN\u2019s":[36],"footprint":[38],"maximize":[40],"its":[41],"execution":[42],"efficiency,":[43],"with":[44,165,179,194,205],"negligible":[45],"end-to-end":[46,131,195],"precision":[47],"degradation.":[48],"In":[49],"this":[50],"work,":[51],"we":[52,103,124,191],"present":[53],"novel":[55],"hardware":[56],"software":[58],"stack":[59],"for":[60,129],"energy-efficient":[61],"inference":[62],"mixed-precision":[64,86,101],"Quantized":[65],"Networks":[67],"(QNNs).":[68],"We":[69],"introduce":[70],"Flex-V,":[71],"processor":[73],"based":[74],"the":[76,92,96,107,130,156,182,186,189],"RISC-V":[77],"Instruction":[78],"Set":[79],"Architecture":[80],"(ISA)":[81],"that":[82],"features":[83],"fused":[84],"Mac&Load":[85],"dot":[87],"product":[88],"instructions;":[89],"avoid":[91],"exponential":[93],"growth":[94],"encoding":[97],"space":[98],"due":[99],"variants,":[102],"encode":[104],"formats":[105],"into":[106,115],"Control-Status":[108],"Registers":[109],"(CSRs).":[110],"Flex-V":[111],"core":[112],"is":[113],"integrated":[114],"tightly-coupled":[117],"cluster":[118],"eight":[120],"processors;":[121],"in":[122,159],"addition,":[123],"provide":[125],"full":[127],"framework":[128],"deployment":[132,143],"DNNs":[134],"including":[135],"compiler,":[137],"optimized":[138],"libraries,":[139],"memory-aware":[142],"flow.":[144],"Our":[145],"results":[146],"show":[147],"up":[148,166],"91.5":[150],"MAC/cycle":[151],"3.26":[153],"TOPS/W":[154],"cluster,":[157],"implemented":[158],"commercial":[161],"22nm":[162],"FDX":[163],"technology,":[164],"$":[168,201],"8.5":[169],"\\times$":[170,204],"speed-up,":[171],"an":[173],"area":[174],"overhead":[175],"only":[177],"5.6%":[178],"respect":[180,206],"baseline.":[183],"To":[184],"demonstrate":[185],"capabilities":[187],"architecture,":[190],"benchmark":[192],"it":[193],"real-life":[196],"QNNs,":[197],"improving":[198],"performance":[199],"by":[200],"2":[202],"\\times-2.5":[203],"existing":[208],"solutions":[209],"using":[210],"fully":[211],"flexible":[212],"programmable":[213],"processors.":[214]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":4},{"year":2023,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}