{"id":"https://openalex.org/W4405946671","doi":"https://doi.org/10.1007/s42514-024-00208-9","title":"Convergence-aware operator-wise mixed-precision training","display_name":"Convergence-aware operator-wise mixed-precision training","publication_year":2024,"publication_date":"2024-12-31","ids":{"openalex":"https://openalex.org/W4405946671","doi":"https://doi.org/10.1007/s42514-024-00208-9"},"language":"en","primary_location":{"id":"doi:10.1007/s42514-024-00208-9","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s42514-024-00208-9","pdf_url":"https://link.springer.com/content/pdf/10.1007/s42514-024-00208-9.pdf","source":{"id":"https://openalex.org/S4210190911","display_name":"CCF Transactions on High Performance Computing","issn_l":"2524-4922","issn":["2524-4922","2524-4930"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"CCF Transactions on High Performance Computing","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://link.springer.com/content/pdf/10.1007/s42514-024-00208-9.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5069709187","display_name":"Wenhao Dai","orcid":null},"institutions":[{"id":"https://openalex.org/I204553293","display_name":"China University of Petroleum, Beijing","ror":"https://ror.org/041qf4r12","country_code":"CN","type":"education","lineage":["https://openalex.org/I204553293"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Wenhao Dai","raw_affiliation_strings":["SSSLab, Department of CST, China University of Petroleum-Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"SSSLab, Department of CST, China University of Petroleum-Beijing, Beijing, China","institution_ids":["https://openalex.org/I204553293"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079189930","display_name":"Ziyi Jia","orcid":"https://orcid.org/0009-0008-8949-4794"},"institutions":[{"id":"https://openalex.org/I204553293","display_name":"China University of Petroleum, Beijing","ror":"https://ror.org/041qf4r12","country_code":"CN","type":"education","lineage":["https://openalex.org/I204553293"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ziyi Jia","raw_affiliation_strings":["SSSLab, Department of CST, China University of Petroleum-Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"SSSLab, Department of CST, China University of Petroleum-Beijing, Beijing, China","institution_ids":["https://openalex.org/I204553293"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078743112","display_name":"Yanqin Bai","orcid":"https://orcid.org/0000-0002-6038-0805"},"institutions":[{"id":"https://openalex.org/I204553293","display_name":"China University of Petroleum, Beijing","ror":"https://ror.org/041qf4r12","country_code":"CN","type":"education","lineage":["https://openalex.org/I204553293"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuesi Bai","raw_affiliation_strings":["SSSLab, Department of CST, China University of Petroleum-Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"SSSLab, Department of CST, China University of Petroleum-Beijing, Beijing, China","institution_ids":["https://openalex.org/I204553293"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5060999547","display_name":"Qingxiao Sun","orcid":"https://orcid.org/0000-0003-2927-362X"},"institutions":[{"id":"https://openalex.org/I204553293","display_name":"China University of Petroleum, Beijing","ror":"https://ror.org/041qf4r12","country_code":"CN","type":"education","lineage":["https://openalex.org/I204553293"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qingxiao Sun","raw_affiliation_strings":["SSSLab, Department of CST, China University of Petroleum-Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"SSSLab, Department of CST, China University of Petroleum-Beijing, Beijing, China","institution_ids":["https://openalex.org/I204553293"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5069709187"],"corresponding_institution_ids":["https://openalex.org/I204553293"],"apc_list":{"value":2190,"currency":"EUR","value_usd":2790},"apc_paid":{"value":2190,"currency":"EUR","value_usd":2790},"fwci":6.9571,"has_fulltext":true,"cited_by_count":5,"citation_normalized_percentile":{"value":0.96432039,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":98},"biblio":{"volume":"7","issue":"1","first_page":"43","last_page":"57"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10191","display_name":"Robotics and Sensor-Based Localization","score":0.9972000122070312,"subfield":{"id":"https://openalex.org/subfields/2202","display_name":"Aerospace Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10191","display_name":"Robotics and Sensor-Based Localization","score":0.9972000122070312,"subfield":{"id":"https://openalex.org/subfields/2202","display_name":"Aerospace 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/T11325","display_name":"Inertial Sensor and Navigation","score":0.996399998664856,"subfield":{"id":"https://openalex.org/subfields/2202","display_name":"Aerospace 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/T10888","display_name":"Augmented Reality Applications","score":0.9853000044822693,"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/convergence","display_name":"Convergence (economics)","score":0.7036047577857971},{"id":"https://openalex.org/keywords/operator","display_name":"Operator (biology)","score":0.6162515878677368},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5921047329902649},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.5510193109512329},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4081438481807709},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3721488416194916},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.35206103324890137},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.33700406551361084},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.26316627860069275},{"id":"https://openalex.org/keywords/geography","display_name":"Geography","score":0.07562181353569031},{"id":"https://openalex.org/keywords/economics","display_name":"Economics","score":0.06246885657310486}],"concepts":[{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.7036047577857971},{"id":"https://openalex.org/C17020691","wikidata":"https://www.wikidata.org/wiki/Q139677","display_name":"Operator (biology)","level":5,"score":0.6162515878677368},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5921047329902649},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5510193109512329},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4081438481807709},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3721488416194916},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.35206103324890137},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.33700406551361084},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.26316627860069275},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.07562181353569031},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.06246885657310486},{"id":"https://openalex.org/C158448853","wikidata":"https://www.wikidata.org/wiki/Q425218","display_name":"Repressor","level":4,"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/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0},{"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/C86339819","wikidata":"https://www.wikidata.org/wiki/Q407384","display_name":"Transcription factor","level":3,"score":0.0},{"id":"https://openalex.org/C50522688","wikidata":"https://www.wikidata.org/wiki/Q189833","display_name":"Economic growth","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1007/s42514-024-00208-9","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s42514-024-00208-9","pdf_url":"https://link.springer.com/content/pdf/10.1007/s42514-024-00208-9.pdf","source":{"id":"https://openalex.org/S4210190911","display_name":"CCF Transactions on High Performance Computing","issn_l":"2524-4922","issn":["2524-4922","2524-4930"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"CCF Transactions on High Performance Computing","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1007/s42514-024-00208-9","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s42514-024-00208-9","pdf_url":"https://link.springer.com/content/pdf/10.1007/s42514-024-00208-9.pdf","source":{"id":"https://openalex.org/S4210190911","display_name":"CCF Transactions on High Performance Computing","issn_l":"2524-4922","issn":["2524-4922","2524-4930"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"CCF Transactions on High Performance Computing","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2087396116","display_name":null,"funder_award_id":"China","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G2376276132","display_name":null,"funder_award_id":"China","funder_id":"https://openalex.org/F4320335787","funder_display_name":"Fundamental Research Funds for the Central Universities"},{"id":"https://openalex.org/G3085993365","display_name":null,"funder_award_id":"(Grant No.","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G3317480652","display_name":null,"funder_award_id":"Science","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G5911307808","display_name":null,"funder_award_id":"2462023YJRC023","funder_id":"https://openalex.org/F4320335787","funder_display_name":"Fundamental Research Funds for the Central Universities"},{"id":"https://openalex.org/G5994120800","display_name":null,"funder_award_id":"Natural","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G7726157001","display_name":null,"funder_award_id":"Grant No.","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G8484223723","display_name":null,"funder_award_id":"62402525","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G8951484681","display_name":null,"funder_award_id":"Grant","funder_id":"https://openalex.org/F4320335787","funder_display_name":"Fundamental Research Funds for the Central Universities"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320335787","display_name":"Fundamental Research Funds for the Central Universities","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4405946671.pdf","grobid_xml":"https://content.openalex.org/works/W4405946671.grobid-xml"},"referenced_works_count":43,"referenced_works":["https://openalex.org/W1975818677","https://openalex.org/W1977146902","https://openalex.org/W2117539524","https://openalex.org/W2137994790","https://openalex.org/W2610467942","https://openalex.org/W2618530766","https://openalex.org/W2765439756","https://openalex.org/W2801035805","https://openalex.org/W2893007245","https://openalex.org/W2895305554","https://openalex.org/W2914973130","https://openalex.org/W2942454403","https://openalex.org/W2951815798","https://openalex.org/W2952851876","https://openalex.org/W2963323070","https://openalex.org/W2970971581","https://openalex.org/W2982479999","https://openalex.org/W2997958863","https://openalex.org/W3034887213","https://openalex.org/W3090334624","https://openalex.org/W3096831136","https://openalex.org/W3118608800","https://openalex.org/W3126212707","https://openalex.org/W3132185085","https://openalex.org/W3137147200","https://openalex.org/W3187908937","https://openalex.org/W3205436637","https://openalex.org/W4283368428","https://openalex.org/W4293024135","https://openalex.org/W4312968147","https://openalex.org/W4318541553","https://openalex.org/W4360831846","https://openalex.org/W4381327277","https://openalex.org/W4384705431","https://openalex.org/W4387757648","https://openalex.org/W4388665000","https://openalex.org/W4391952402","https://openalex.org/W4393407124","https://openalex.org/W4399282175","https://openalex.org/W6601487527","https://openalex.org/W6604186633","https://openalex.org/W6605494067","https://openalex.org/W6713134421"],"related_works":["https://openalex.org/W230091440","https://openalex.org/W2233261550","https://openalex.org/W2810751659","https://openalex.org/W258997015","https://openalex.org/W2997094352","https://openalex.org/W3216976533","https://openalex.org/W100620283","https://openalex.org/W2495260952","https://openalex.org/W4366179611","https://openalex.org/W2996078371"],"abstract_inverted_index":{"Abstract":[0],"With":[1],"the":[2,78,88,128],"support":[3],"of":[4,61,81,148,166],"more":[5],"precision":[6,70,82,107],"formats":[7,26],"in":[8],"emerging":[9],"hardware":[10],"architectures,":[11],"mixed-precision":[12,52,132],"has":[13],"become":[14],"a":[15,59,96,144],"popular":[16],"approach":[17],"to":[18,32,58,114,127,164],"accelerate":[19],"deep":[20],"learning":[21],"(DL)":[22],"training.":[23,103],"Applying":[24],"low-precision":[25],"such":[27],"as":[28,50],"FP16":[29],"and":[30,47,54,77,112,117],"BF16":[31],"neural":[33,62],"operators":[34],"can":[35],"save":[36],"GPU":[37,139,167],"memory":[38],"while":[39,160],"improving":[40],"bandwidth.":[41],"However,":[42],"DL":[43],"frameworks":[44],"use":[45],"black":[46],"white":[48],"lists":[49],"default":[51],"selections":[53],"cannot":[55],"flexibly":[56],"adapt":[57],"variety":[60],"networks.":[63],"In":[64],"addition,":[65],"existing":[66],"work":[67],"on":[68,110,137],"automatic":[69],"adjustment":[71,108],"does":[72],"not":[73],"consider":[74],"model":[75],"convergence,":[76],"decision":[79],"cost":[80],"selection":[83],"is":[84],"high.":[85],"To":[86],"address":[87],"above":[89],"problems,":[90],"this":[91],"paper":[92],"proposes":[93],"CoMP":[94,104,122,142],",":[95],"non-intrusive":[97],"framework":[98],"for":[99],"Convergence-aware":[100],"operator-wise":[101,131],"Mixed-precision":[102],"uses":[105],"two-stage":[106],"based":[109],"epochs":[111],"batches":[113],"ensure":[115],"convergence":[116],"performance":[118,146],"respectively.":[119],"After":[120],"that,":[121],"performs":[123],"subsequent":[124],"training":[125],"according":[126],"searched":[129],"optimal":[130],"plan.":[133],"The":[134],"experimental":[135],"results":[136],"A100":[138],"show":[140],"that":[141],"achieves":[143],"maximum":[145],"speedup":[147],"1.15":[149],"$$\\times$$":[150],"<mml:math":[151],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\">":[152],"<mml:mo>\u00d7</mml:mo>":[153],"</mml:math>":[154],"compared":[155],"with":[156],"PyTorch":[157],"AMP":[158],"implementation,":[159],"also":[161],"saving":[162],"up":[163],"29.81%":[165],"memory.":[168]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":4}],"updated_date":"2026-04-13T07:58:08.660418","created_date":"2025-10-10T00:00:00"}