{"id":"https://openalex.org/W4378800959","doi":"https://doi.org/10.1145/3583781.3590269","title":"Low-Cost Multiple-Precision Multiplication Unit Design For Deep Learning","display_name":"Low-Cost Multiple-Precision Multiplication Unit Design For Deep Learning","publication_year":2023,"publication_date":"2023-05-31","ids":{"openalex":"https://openalex.org/W4378800959","doi":"https://doi.org/10.1145/3583781.3590269"},"language":"en","primary_location":{"id":"doi:10.1145/3583781.3590269","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3583781.3590269","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Great Lakes Symposium on VLSI 2023","raw_type":"proceedings-article"},"type":"article","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/A5101985792","display_name":"Jing Zhang","orcid":"https://orcid.org/0009-0002-7087-2528"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jing Zhang","raw_affiliation_strings":["National University of Defense Technology, Changsha, China"],"raw_orcid":"https://orcid.org/0009-0002-7087-2528","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069512300","display_name":"Libo Huang","orcid":"https://orcid.org/0000-0002-8307-6742"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Libo Huang","raw_affiliation_strings":["National University of Defense Technology, Changsha, China"],"raw_orcid":"https://orcid.org/0000-0002-8307-6742","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019476529","display_name":"Hongbing Tan","orcid":"https://orcid.org/0000-0003-4184-4173"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hongbing Tan","raw_affiliation_strings":["National University of Defense Technology, Changsha, China"],"raw_orcid":"https://orcid.org/0000-0003-4184-4173","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054046358","display_name":"Ling Yang","orcid":"https://orcid.org/0000-0002-0838-0066"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ling Yang","raw_affiliation_strings":["National University of Defense Technology, Changsha, China"],"raw_orcid":"https://orcid.org/0000-0002-0838-0066","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102973587","display_name":"Zhong Zheng","orcid":"https://orcid.org/0009-0009-5326-4927"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhong Zheng","raw_affiliation_strings":["National University of Defense Technology, Changsha, China"],"raw_orcid":"https://orcid.org/0009-0009-5326-4927","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5113537087","display_name":"Qianming Yang","orcid":null},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qianming Yang","raw_affiliation_strings":["National University of Defense Technology, Changsha, China"],"raw_orcid":"https://orcid.org/0000-0002-8796-6187","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.7389,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.74375021,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"9","last_page":"14"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11697","display_name":"Numerical Methods and Algorithms","score":0.9994999766349792,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T11697","display_name":"Numerical Methods and Algorithms","score":0.9994999766349792,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9975000023841858,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T11693","display_name":"Cryptography and Residue Arithmetic","score":0.992900013923645,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.801674485206604},{"id":"https://openalex.org/keywords/multiplication","display_name":"Multiplication (music)","score":0.6335214376449585},{"id":"https://openalex.org/keywords/single-precision-floating-point-format","display_name":"Single-precision floating-point format","score":0.5420878529548645},{"id":"https://openalex.org/keywords/floating-point","display_name":"Floating point","score":0.5133432745933533},{"id":"https://openalex.org/keywords/low-latency","display_name":"Low latency (capital markets)","score":0.5081243515014648},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.497088223695755},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.49140942096710205},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.48454228043556213},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.47768884897232056},{"id":"https://openalex.org/keywords/floating-point-unit","display_name":"Floating-point unit","score":0.46807989478111267},{"id":"https://openalex.org/keywords/multiplication-algorithm","display_name":"Multiplication algorithm","score":0.4618918001651764},{"id":"https://openalex.org/keywords/multiplier","display_name":"Multiplier (economics)","score":0.4351067543029785},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.4258642792701721},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.33215153217315674},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.33129000663757324},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3170170783996582},{"id":"https://openalex.org/keywords/arithmetic","display_name":"Arithmetic","score":0.30773311853408813},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2776651978492737},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.11375859379768372}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.801674485206604},{"id":"https://openalex.org/C2780595030","wikidata":"https://www.wikidata.org/wiki/Q3860309","display_name":"Multiplication (music)","level":2,"score":0.6335214376449585},{"id":"https://openalex.org/C133095886","wikidata":"https://www.wikidata.org/wiki/Q1307173","display_name":"Single-precision floating-point format","level":3,"score":0.5420878529548645},{"id":"https://openalex.org/C84211073","wikidata":"https://www.wikidata.org/wiki/Q117879","display_name":"Floating point","level":2,"score":0.5133432745933533},{"id":"https://openalex.org/C46637626","wikidata":"https://www.wikidata.org/wiki/Q6693015","display_name":"Low latency (capital markets)","level":2,"score":0.5081243515014648},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.497088223695755},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.49140942096710205},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.48454228043556213},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.47768884897232056},{"id":"https://openalex.org/C110305270","wikidata":"https://www.wikidata.org/wiki/Q733507","display_name":"Floating-point unit","level":3,"score":0.46807989478111267},{"id":"https://openalex.org/C201290732","wikidata":"https://www.wikidata.org/wiki/Q130762","display_name":"Multiplication algorithm","level":3,"score":0.4618918001651764},{"id":"https://openalex.org/C124584101","wikidata":"https://www.wikidata.org/wiki/Q1053266","display_name":"Multiplier (economics)","level":2,"score":0.4351067543029785},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.4258642792701721},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.33215153217315674},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.33129000663757324},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3170170783996582},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.30773311853408813},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2776651978492737},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11375859379768372},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","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},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","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/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3583781.3590269","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3583781.3590269","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Great Lakes Symposium on VLSI 2023","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":19,"referenced_works":["https://openalex.org/W2026445983","https://openalex.org/W2207050309","https://openalex.org/W2618530766","https://openalex.org/W2798490684","https://openalex.org/W2801795932","https://openalex.org/W2914973130","https://openalex.org/W2919115771","https://openalex.org/W2947629474","https://openalex.org/W2997889117","https://openalex.org/W3033696290","https://openalex.org/W3163798119","https://openalex.org/W3187235458","https://openalex.org/W3213775513","https://openalex.org/W4236302272","https://openalex.org/W4285055485","https://openalex.org/W4301045096","https://openalex.org/W4312872667","https://openalex.org/W6687483927","https://openalex.org/W6843734122"],"related_works":["https://openalex.org/W3215589575","https://openalex.org/W2116803521","https://openalex.org/W2797902698","https://openalex.org/W2363539709","https://openalex.org/W1564887326","https://openalex.org/W2541658314","https://openalex.org/W2000092506","https://openalex.org/W2169016399","https://openalex.org/W1589067093","https://openalex.org/W2596474508"],"abstract_inverted_index":{"Low-precision":[0],"formats":[1,155],"have":[2],"been":[3],"proposed":[4,29],"and":[5,15,52,88,104,120,160],"applied":[6],"to":[7,11,101,129],"deep":[8,26,97,145],"learning":[9,98],"algorithms":[10],"speed":[12],"up":[13],"training":[14],"inference.":[16],"This":[17],"paper":[18],"proposes":[19],"a":[20,69,78,125],"novel":[21],"multiple-precision":[22,141],"multiplication":[23,76,108],"unit(MU)":[24],"for":[25,36,144],"learning.":[27],"The":[28,45,83],"MU":[30,46,122,148],"supports":[31],"four":[32,49,62],"types":[33,152],"of":[34,71,80,86,94,116,153],"precision":[35,111],"floating-point(FP)":[37],"numbers-FP8-E4M3,":[38],"FP8-E5M2,":[39],"FP16,":[40],"FP32-and":[41],"8-bit":[42],"fixed-point(FIX)":[43],"numbers.":[44],"can":[47,90,112,123,149],"execute":[48],"parallel":[50,54,63],"FP8":[51,89],"eight":[53],"FIX8":[55,87,165],"multiplications":[56,65],"simultaneously":[57],"in":[58,109,133],"one":[59,74,81],"cycle,":[60],"or":[61,73],"FP16":[64],"fully":[66],"pipelined":[67],"with":[68,77,138,166],"latency":[70,79],"one,":[72],"FP32":[75],"cycle.":[82],"simultaneous":[84],"execution":[85],"meet":[91],"the":[92,95,102,117,121,139],"requirements":[93],"specific":[96],"algorithms.":[99],"Thanks":[100],"low-precision-combination(LPC)":[103],"vectorization":[105],"design":[106],"method,":[107],"any":[110],"get":[113],"100%":[114],"utilization":[115],"multiplier":[118],"resources,":[119],"adopt":[124],"lower":[126,157],"clock":[127],"delay":[128],"achieve":[130],"better":[131],"performance":[132],"all":[134],"data":[135],"types.":[136],"Compared":[137],"existing":[140],"units":[142],"designed":[143],"learning,":[146],"this":[147],"support":[150],"more":[151],"low-precision":[154],"by":[156],"area":[158],"overhead;":[159],"exhibits":[161],"higher":[162],"throughput":[163],"at":[164,167],"least":[168],"8\u00d7":[169],"improvement.":[170]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1}],"updated_date":"2026-06-19T17:40:00.097472","created_date":"2025-10-10T00:00:00"}