{"id":"https://openalex.org/W3167622379","doi":"https://doi.org/10.1145/3447548.3467287","title":"A Stagewise Hyperparameter Scheduler to Improve Generalization","display_name":"A Stagewise Hyperparameter Scheduler to Improve Generalization","publication_year":2021,"publication_date":"2021-08-12","ids":{"openalex":"https://openalex.org/W3167622379","doi":"https://doi.org/10.1145/3447548.3467287","mag":"3167622379"},"language":"en","primary_location":{"id":"doi:10.1145/3447548.3467287","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3447548.3467287","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining","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/A5021218766","display_name":"Jianhui Sun","orcid":"https://orcid.org/0000-0003-0032-3646"},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Jianhui Sun","raw_affiliation_strings":["University of Virginia, Charlottesville, VA, USA"],"affiliations":[{"raw_affiliation_string":"University of Virginia, Charlottesville, VA, USA","institution_ids":["https://openalex.org/I51556381"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074173740","display_name":"Ying Yang","orcid":"https://orcid.org/0000-0003-1602-1212"},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan\u2013Ann Arbor","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ying Yang","raw_affiliation_strings":["University of Michigan, Ann Arbor, MI, USA"],"affiliations":[{"raw_affiliation_string":"University of Michigan, Ann Arbor, MI, USA","institution_ids":["https://openalex.org/I27837315"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048238468","display_name":"Guangxu Xun","orcid":"https://orcid.org/0000-0002-7657-4305"},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Guangxu Xun","raw_affiliation_strings":["University of Virginia, Charlottesville, VA, USA"],"affiliations":[{"raw_affiliation_string":"University of Virginia, Charlottesville, VA, USA","institution_ids":["https://openalex.org/I51556381"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5013588572","display_name":"Aidong Zhang","orcid":"https://orcid.org/0000-0001-9723-3246"},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Aidong Zhang","raw_affiliation_strings":["University of Virginia, Charlottesville, VA, USA"],"affiliations":[{"raw_affiliation_string":"University of Virginia, Charlottesville, VA, USA","institution_ids":["https://openalex.org/I51556381"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5021218766"],"corresponding_institution_ids":["https://openalex.org/I51556381"],"apc_list":null,"apc_paid":null,"fwci":1.0877,"has_fulltext":false,"cited_by_count":8,"citation_normalized_percentile":{"value":0.81507306,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1530","last_page":"1540"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9998000264167786,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9998000264167786,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.9991000294685364,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.9990000128746033,"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/hyperparameter","display_name":"Hyperparameter","score":0.9192536473274231},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6485186815261841},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.5294287800788879},{"id":"https://openalex.org/keywords/momentum","display_name":"Momentum (technical analysis)","score":0.48307469487190247},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.46305060386657715},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.4358682334423065},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.4205951690673828},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3973901867866516},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3837466239929199},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3511563539505005},{"id":"https://openalex.org/keywords/applied-mathematics","display_name":"Applied mathematics","score":0.3454853892326355},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.2779286503791809},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.18935823440551758}],"concepts":[{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.9192536473274231},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6485186815261841},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.5294287800788879},{"id":"https://openalex.org/C60718061","wikidata":"https://www.wikidata.org/wiki/Q1414747","display_name":"Momentum (technical analysis)","level":2,"score":0.48307469487190247},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.46305060386657715},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.4358682334423065},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.4205951690673828},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3973901867866516},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3837466239929199},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3511563539505005},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.3454853892326355},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2779286503791809},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.18935823440551758},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","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/C10138342","wikidata":"https://www.wikidata.org/wiki/Q43015","display_name":"Finance","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3447548.3467287","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3447548.3467287","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G5628663453","display_name":null,"funder_award_id":"IIS-2008208, IIS-1934600, IIS-1938167, IIS-1955151","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":35,"referenced_works":["https://openalex.org/W1806891645","https://openalex.org/W1988720110","https://openalex.org/W2014384147","https://openalex.org/W2097998348","https://openalex.org/W2163605009","https://openalex.org/W2588845097","https://openalex.org/W2605488176","https://openalex.org/W2625885504","https://openalex.org/W2726042758","https://openalex.org/W2752513799","https://openalex.org/W2808402975","https://openalex.org/W2883109957","https://openalex.org/W2896222303","https://openalex.org/W2913357890","https://openalex.org/W2914304175","https://openalex.org/W2920882081","https://openalex.org/W2936995161","https://openalex.org/W2941910275","https://openalex.org/W2944572616","https://openalex.org/W2951538941","https://openalex.org/W2962915600","https://openalex.org/W2963243933","https://openalex.org/W2963433607","https://openalex.org/W2964054038","https://openalex.org/W2964102336","https://openalex.org/W2990719156","https://openalex.org/W2998508934","https://openalex.org/W3007957341","https://openalex.org/W3008463948","https://openalex.org/W3012252768","https://openalex.org/W3042549348","https://openalex.org/W3080994088","https://openalex.org/W3139294691","https://openalex.org/W4255556797","https://openalex.org/W6674385629"],"related_works":["https://openalex.org/W4295309597","https://openalex.org/W4323894855","https://openalex.org/W3204187020","https://openalex.org/W4210794429","https://openalex.org/W4223456145","https://openalex.org/W4309113015","https://openalex.org/W4214718235","https://openalex.org/W3108405844","https://openalex.org/W4210831666","https://openalex.org/W4360898066"],"abstract_inverted_index":{"Stochastic":[0],"gradient":[1,17],"descent":[2],"(SGD)":[3],"augmented":[4],"with":[5],"various":[6],"momentum":[7,12,43,81,93,125,144],"variants":[8,82,145],"(e.g.":[9,121,150],"heavy":[10],"ball":[11],"(SHB)":[13],"and":[14,61,95,98,112,172,174,223],"Nesterov's":[15],"accelerated":[16],"(NAG))":[18],"has":[19],"been":[20],"the":[21,29,34,59,88,100,207,220],"default":[22],"optimizer":[23],"for":[24,66,212],"many":[25],"learning":[26,102],"tasks.":[27],"Tuning":[28],"optimizer's":[30],"hyperparameters":[31,55,120,167],"is":[32,64,108,134],"arguably":[33],"most":[35],"time-consuming":[36],"part":[37],"of":[38,79,91,143,209],"model":[39],"training.":[40],"Many":[41],"new":[42],"variants,":[44],"despite":[45],"their":[46],"empirical":[47],"advantage":[48],"over":[49],"classical":[50],"SHB/NAG,":[51],"introduce":[52],"even":[53],"more":[54],"to":[56,73,83,118],"tune.":[57],"Automating":[58],"tedious":[60],"error-prone":[62],"tuning":[63,181,184,199,221],"essential":[65],"AutoML.":[67],"This":[68],"paper":[69],"focuses":[70],"on":[71,157],"how":[72],"efficiently":[74],"tune":[75],"a":[76,135,140,191],"large":[77,110,141],"class":[78],"multistage":[80,162,210],"improve":[84,202],"generalization.":[85,203],"We":[86,194,204],"use":[87],"general":[89,213],"formulation":[90],"quasi-hyperbolic":[92],"(QHM)":[94],"extend":[96],"\"constant":[97],"drop'',":[99],"widespread":[101],"rate":[103],"\u03b1":[104,107],"scheduler":[105],"where":[106],"set":[109],"initially":[111],"then":[113],"dropped":[114],"every":[115],"few":[116],"epochs,":[117],"other":[119],"batch":[122],"size":[123],"b,":[124,170],"parameter":[126],"\u03b2,":[127,171],"instant":[128],"discount":[129],"factor":[130],"\u03bd).":[131],"Multistage":[132],"QHM":[133,163,211],"unified":[136],"framework":[137],"which":[138],"covers":[139],"family":[142],"as":[146,169],"its":[147],"special":[148],"cases":[149],"vanilla":[151],"SGD/SHB/NAG).":[152],"Existing":[153],"works":[154],"mainly":[155],"focus":[156],"scheduling":[158],"\u03b1's":[159],"decay,":[160],"while":[161],"allows":[164],"additional":[165],"varying":[166],"such":[168],"\u03bd,":[173],"demonstrates":[175],"better":[176],"generalization":[177],"ability":[178],"than":[179,190],"only":[180],"\u03b1.":[182],"Our":[183,217],"strategies":[185,200,218],"have":[186],"rigorous":[187],"justifications":[188],"rather":[189],"blind":[192],"trial-and-error.":[193],"theoretically":[195],"prove":[196],"why":[197],"our":[198],"could":[201],"also":[205],"show":[206],"convergence":[208],"nonconvex":[214],"objective":[215],"functions.":[216],"simplify":[219],"process":[222],"beat":[224],"competitive":[225],"optimizers":[226],"in":[227],"test":[228],"accuracy":[229],"empirically.":[230]},"counts_by_year":[{"year":2024,"cited_by_count":4},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}