{"id":"https://openalex.org/W2915390278","doi":"https://doi.org/10.1109/cahpc.2018.8645923","title":"A Case Study on Optimizing Accurate Half Precision Average","display_name":"A Case Study on Optimizing Accurate Half Precision Average","publication_year":2018,"publication_date":"2018-09-01","ids":{"openalex":"https://openalex.org/W2915390278","doi":"https://doi.org/10.1109/cahpc.2018.8645923","mag":"2915390278"},"language":"en","primary_location":{"id":"doi:10.1109/cahpc.2018.8645923","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cahpc.2018.8645923","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 30th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD)","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/A5017754542","display_name":"Kenny Peou","orcid":null},"institutions":[{"id":"https://openalex.org/I1294671590","display_name":"Centre National de la Recherche Scientifique","ror":"https://ror.org/02feahw73","country_code":"FR","type":"funder","lineage":["https://openalex.org/I1294671590"]},{"id":"https://openalex.org/I4210144804","display_name":"Laboratoire de Recherche en Informatique","ror":"https://ror.org/04e3ktk27","country_code":"FR","type":"facility","lineage":["https://openalex.org/I102197404","https://openalex.org/I1294671590","https://openalex.org/I1326498283","https://openalex.org/I4210144804","https://openalex.org/I4210159245"]},{"id":"https://openalex.org/I277688954","display_name":"Universit\u00e9 Paris-Saclay","ror":"https://ror.org/03xjwb503","country_code":"FR","type":"education","lineage":["https://openalex.org/I277688954"]}],"countries":["FR"],"is_corresponding":true,"raw_author_name":"Kenny Peou","raw_affiliation_strings":["LRI UMR CNRS 8623, Universit\u00e9 Paris-Saclay"],"affiliations":[{"raw_affiliation_string":"LRI UMR CNRS 8623, Universit\u00e9 Paris-Saclay","institution_ids":["https://openalex.org/I277688954","https://openalex.org/I4210144804","https://openalex.org/I1294671590"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048673635","display_name":"Alan L. Kelly","orcid":"https://orcid.org/0000-0003-2759-1587"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Alan Kelly","raw_affiliation_strings":["NUMSCALE"],"affiliations":[{"raw_affiliation_string":"NUMSCALE","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035522197","display_name":"Jo\u00ebl Falcou","orcid":"https://orcid.org/0000-0001-5380-7375"},"institutions":[{"id":"https://openalex.org/I277688954","display_name":"Universit\u00e9 Paris-Saclay","ror":"https://ror.org/03xjwb503","country_code":"FR","type":"education","lineage":["https://openalex.org/I277688954"]},{"id":"https://openalex.org/I4210144804","display_name":"Laboratoire de Recherche en Informatique","ror":"https://ror.org/04e3ktk27","country_code":"FR","type":"facility","lineage":["https://openalex.org/I102197404","https://openalex.org/I1294671590","https://openalex.org/I1326498283","https://openalex.org/I4210144804","https://openalex.org/I4210159245"]},{"id":"https://openalex.org/I1294671590","display_name":"Centre National de la Recherche Scientifique","ror":"https://ror.org/02feahw73","country_code":"FR","type":"funder","lineage":["https://openalex.org/I1294671590"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Joel Falcou","raw_affiliation_strings":["LRI UMR CNRS 8623, Universit\u00e9 Paris-Saclay"],"affiliations":[{"raw_affiliation_string":"LRI UMR CNRS 8623, Universit\u00e9 Paris-Saclay","institution_ids":["https://openalex.org/I277688954","https://openalex.org/I4210144804","https://openalex.org/I1294671590"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5036547514","display_name":"C\u00e9cile Germain","orcid":"https://orcid.org/0000-0002-9764-9783"},"institutions":[{"id":"https://openalex.org/I4210144804","display_name":"Laboratoire de Recherche en Informatique","ror":"https://ror.org/04e3ktk27","country_code":"FR","type":"facility","lineage":["https://openalex.org/I102197404","https://openalex.org/I1294671590","https://openalex.org/I1326498283","https://openalex.org/I4210144804","https://openalex.org/I4210159245"]},{"id":"https://openalex.org/I1294671590","display_name":"Centre National de la Recherche Scientifique","ror":"https://ror.org/02feahw73","country_code":"FR","type":"funder","lineage":["https://openalex.org/I1294671590"]},{"id":"https://openalex.org/I277688954","display_name":"Universit\u00e9 Paris-Saclay","ror":"https://ror.org/03xjwb503","country_code":"FR","type":"education","lineage":["https://openalex.org/I277688954"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Cecile Germain","raw_affiliation_strings":["LRI UMR CNRS 8623, Universit\u00e9 Paris-Saclay"],"affiliations":[{"raw_affiliation_string":"LRI UMR CNRS 8623, Universit\u00e9 Paris-Saclay","institution_ids":["https://openalex.org/I277688954","https://openalex.org/I4210144804","https://openalex.org/I1294671590"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5017754542"],"corresponding_institution_ids":["https://openalex.org/I1294671590","https://openalex.org/I277688954","https://openalex.org/I4210144804"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.22075907,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":94},"biblio":{"volume":"4","issue":null,"first_page":"356","last_page":"363"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11697","display_name":"Numerical Methods and Algorithms","score":1.0,"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":1.0,"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/T11435","display_name":"Polynomial and algebraic computation","score":0.9909999966621399,"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/T11693","display_name":"Cryptography and Residue Arithmetic","score":0.9890000224113464,"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/simd","display_name":"SIMD","score":0.8766293525695801},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.7753006219863892},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.7513453960418701},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7301648855209351},{"id":"https://openalex.org/keywords/floating-point","display_name":"Floating point","score":0.7210046052932739},{"id":"https://openalex.org/keywords/double-precision-floating-point-format","display_name":"Double-precision floating-point format","score":0.6069151163101196},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.5727050304412842},{"id":"https://openalex.org/keywords/single-precision-floating-point-format","display_name":"Single-precision floating-point format","score":0.5686663389205933},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5332678556442261},{"id":"https://openalex.org/keywords/ieee-floating-point","display_name":"IEEE floating point","score":0.48087504506111145},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.4512483477592468}],"concepts":[{"id":"https://openalex.org/C150552126","wikidata":"https://www.wikidata.org/wiki/Q339387","display_name":"SIMD","level":2,"score":0.8766293525695801},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.7753006219863892},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.7513453960418701},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7301648855209351},{"id":"https://openalex.org/C84211073","wikidata":"https://www.wikidata.org/wiki/Q117879","display_name":"Floating point","level":2,"score":0.7210046052932739},{"id":"https://openalex.org/C35912277","wikidata":"https://www.wikidata.org/wiki/Q1243369","display_name":"Double-precision floating-point format","level":3,"score":0.6069151163101196},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5727050304412842},{"id":"https://openalex.org/C133095886","wikidata":"https://www.wikidata.org/wiki/Q1307173","display_name":"Single-precision floating-point format","level":3,"score":0.5686663389205933},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5332678556442261},{"id":"https://openalex.org/C137231763","wikidata":"https://www.wikidata.org/wiki/Q828287","display_name":"IEEE floating point","level":3,"score":0.48087504506111145},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.4512483477592468},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cahpc.2018.8645923","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cahpc.2018.8645923","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 30th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":15,"referenced_works":["https://openalex.org/W605824955","https://openalex.org/W1557855942","https://openalex.org/W1561994953","https://openalex.org/W1841592590","https://openalex.org/W1986624837","https://openalex.org/W2010400056","https://openalex.org/W2013156670","https://openalex.org/W2126840262","https://openalex.org/W2127218421","https://openalex.org/W2164500538","https://openalex.org/W2169004268","https://openalex.org/W2529546376","https://openalex.org/W2963374099","https://openalex.org/W4240170991","https://openalex.org/W6638783484"],"related_works":["https://openalex.org/W1564887326","https://openalex.org/W3215589575","https://openalex.org/W2587538765","https://openalex.org/W2000092506","https://openalex.org/W2596474508","https://openalex.org/W2169016399","https://openalex.org/W2730658480","https://openalex.org/W2116803521","https://openalex.org/W1573821047","https://openalex.org/W2565957982"],"abstract_inverted_index":{"In":[0],"this":[1],"work,":[2],"we":[3],"study":[4],"the":[5,15,28,55,61,64,90,96,125],"numerical":[6,106,119],"performance":[7,123],"of":[8,17,20,31,47,63,89,104],"various":[9],"common":[10],"algorithms":[11,69],"used":[12],"to":[13,60,95,124],"calculate":[14],"average":[16,91],"an":[18,81],"array":[19],"half":[21,56],"precision":[22,107,127],"(FP16)":[23],"floating":[24,98],"point":[25,99],"values.":[26],"While":[27],"current":[29],"generation":[30],"CPUs":[32],"does":[33],"not":[34],"support":[35],"native":[36],"FP16":[37,50,65,87],"arithmetic,":[38],"it":[39,94],"is":[40],"a":[41,45],"planned":[42],"feature":[43],"in":[44,102],"number":[46],"next-generation":[48],"CPUs.":[49],"arithmetic":[51],"was":[52],"emulated":[53],"via":[54],"software":[57],"library.":[58],"Due":[59],"limitations":[62],"data":[66],"type,":[67],"some":[68],"proved":[70],"insufficient":[71],"for":[72],"arrays":[73],"as":[74,76],"small":[75],"100":[77],"elements.":[78],"We":[79,111],"propose":[80],"algorithm":[82,101,115],"that":[83,113],"allows":[84],"numerically":[85],"stable":[86],"computation":[88],"and":[92,108,121],"compare":[93],"naive":[97],"(FP32)":[100],"terms":[103],"both":[105],"runtime":[109],"performance.":[110],"find":[112],"our":[114],"offers":[116],"comparable":[117],"robustness,":[118],"precision,":[120],"SIMD":[122],"higher":[126],"computation.":[128]},"counts_by_year":[{"year":2023,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}
