{"id":"https://openalex.org/W3160259745","doi":"https://doi.org/10.1109/icassp39728.2021.9413928","title":"SapAugment: Learning A Sample Adaptive Policy for Data Augmentation","display_name":"SapAugment: Learning A Sample Adaptive Policy for Data Augmentation","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3160259745","doi":"https://doi.org/10.1109/icassp39728.2021.9413928","mag":"3160259745"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9413928","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9413928","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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/A5062098630","display_name":"Ting-Yao Hu","orcid":null},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Ting-Yao Hu","raw_affiliation_strings":["Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002740493","display_name":"Ashish Shrivastava","orcid":"https://orcid.org/0000-0002-2465-0708"},"institutions":[{"id":"https://openalex.org/I4210107260","display_name":"Apple (United Kingdom)","ror":"https://ror.org/01vpeym60","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210107260"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Ashish Shrivastava","raw_affiliation_strings":["Apple"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple","institution_ids":["https://openalex.org/I4210107260"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066816752","display_name":"Jen-Hao Rick Chang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210107260","display_name":"Apple (United Kingdom)","ror":"https://ror.org/01vpeym60","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210107260"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Jen-Hao Rick Chang","raw_affiliation_strings":["Apple"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple","institution_ids":["https://openalex.org/I4210107260"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064400393","display_name":"Hema Swetha Koppula","orcid":null},"institutions":[{"id":"https://openalex.org/I4210107260","display_name":"Apple (United Kingdom)","ror":"https://ror.org/01vpeym60","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210107260"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Hema Koppula","raw_affiliation_strings":["Apple"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple","institution_ids":["https://openalex.org/I4210107260"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103176031","display_name":"Stefan Braun","orcid":"https://orcid.org/0000-0001-7117-0603"},"institutions":[{"id":"https://openalex.org/I4210107260","display_name":"Apple (United Kingdom)","ror":"https://ror.org/01vpeym60","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210107260"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Stefan Braun","raw_affiliation_strings":["Apple"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple","institution_ids":["https://openalex.org/I4210107260"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057914051","display_name":"Kyuyeon Hwang","orcid":"https://orcid.org/0000-0001-6700-1583"},"institutions":[{"id":"https://openalex.org/I4210107260","display_name":"Apple (United Kingdom)","ror":"https://ror.org/01vpeym60","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210107260"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Kyuyeon Hwang","raw_affiliation_strings":["Apple"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple","institution_ids":["https://openalex.org/I4210107260"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066166549","display_name":"Ozlem Kalinli","orcid":null},"institutions":[{"id":"https://openalex.org/I4210107260","display_name":"Apple (United Kingdom)","ror":"https://ror.org/01vpeym60","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210107260"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Ozlem Kalinli","raw_affiliation_strings":["Apple"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple","institution_ids":["https://openalex.org/I4210107260"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5028613002","display_name":"Oncel Tuzel","orcid":null},"institutions":[{"id":"https://openalex.org/I4210107260","display_name":"Apple (United Kingdom)","ror":"https://ror.org/01vpeym60","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210107260"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Oncel Tuzel","raw_affiliation_strings":["Apple"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple","institution_ids":["https://openalex.org/I4210107260"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5062098630"],"corresponding_institution_ids":["https://openalex.org/I74973139"],"apc_list":null,"apc_paid":null,"fwci":2.0997,"has_fulltext":false,"cited_by_count":17,"citation_normalized_percentile":{"value":0.89195746,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"4040","last_page":"4044"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9918000102043152,"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/T10028","display_name":"Topic Modeling","score":0.9914000034332275,"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/computer-science","display_name":"Computer science","score":0.7179982662200928},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.7029872536659241},{"id":"https://openalex.org/keywords/variance","display_name":"Variance (accounting)","score":0.6203173398971558},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.5824780464172363},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.5248863101005554},{"id":"https://openalex.org/keywords/gaussian-noise","display_name":"Gaussian noise","score":0.5204991698265076},{"id":"https://openalex.org/keywords/sample-size-determination","display_name":"Sample size determination","score":0.5194010734558105},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.5077903866767883},{"id":"https://openalex.org/keywords/noise-reduction","display_name":"Noise reduction","score":0.48552918434143066},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.4632744789123535},{"id":"https://openalex.org/keywords/sample-variance","display_name":"Sample variance","score":0.44032448530197144},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4139208197593689},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4049649238586426},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.37847280502319336},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.26442867517471313},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.18698301911354065}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7179982662200928},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.7029872536659241},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.6203173398971558},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.5824780464172363},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.5248863101005554},{"id":"https://openalex.org/C4199805","wikidata":"https://www.wikidata.org/wiki/Q2725903","display_name":"Gaussian noise","level":2,"score":0.5204991698265076},{"id":"https://openalex.org/C129848803","wikidata":"https://www.wikidata.org/wiki/Q2564360","display_name":"Sample size determination","level":2,"score":0.5194010734558105},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.5077903866767883},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.48552918434143066},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.4632744789123535},{"id":"https://openalex.org/C2993021520","wikidata":"https://www.wikidata.org/wiki/Q175199","display_name":"Sample variance","level":3,"score":0.44032448530197144},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4139208197593689},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4049649238586426},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.37847280502319336},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.26442867517471313},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.18698301911354065},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.0},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"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/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","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/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C121955636","wikidata":"https://www.wikidata.org/wiki/Q4116214","display_name":"Accounting","level":1,"score":0.0},{"id":"https://openalex.org/C43617362","wikidata":"https://www.wikidata.org/wiki/Q170050","display_name":"Chromatography","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp39728.2021.9413928","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9413928","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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":30,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1748012577","https://openalex.org/W1989674786","https://openalex.org/W2099621636","https://openalex.org/W2163605009","https://openalex.org/W2184343439","https://openalex.org/W2407080277","https://openalex.org/W2617258110","https://openalex.org/W2618530766","https://openalex.org/W2783482415","https://openalex.org/W2887528618","https://openalex.org/W2936774411","https://openalex.org/W2949736877","https://openalex.org/W2962780374","https://openalex.org/W2963403868","https://openalex.org/W2970395278","https://openalex.org/W2992308087","https://openalex.org/W2997564604","https://openalex.org/W3015889230","https://openalex.org/W3035682985","https://openalex.org/W4385245566","https://openalex.org/W6637691865","https://openalex.org/W6675409298","https://openalex.org/W6686369942","https://openalex.org/W6713762819","https://openalex.org/W6739901393","https://openalex.org/W6747939174","https://openalex.org/W6753500843","https://openalex.org/W6762334975","https://openalex.org/W6772167297"],"related_works":["https://openalex.org/W4293088233","https://openalex.org/W3152660226","https://openalex.org/W2496077116","https://openalex.org/W2900695351","https://openalex.org/W1570799877","https://openalex.org/W2057598446","https://openalex.org/W2034778756","https://openalex.org/W2744007692","https://openalex.org/W2131347664","https://openalex.org/W2169782520"],"abstract_inverted_index":{"Data":[0],"augmentation":[1,7,77,100,132,175,186,231],"methods":[2,176],"usually":[3],"apply":[4,191],"the":[5,15,25,58,131,136,140,144,170,208,228],"same":[6],"(or":[8],"a":[9,30,34,44,71,75,88,98,106,116,121,149,156,165,178],"mix":[10],"of":[11,108,139,146],"them)":[12],"to":[13,20,56,84,101,105,119,216],"all":[14,39],"training":[16,49,54,92,137],"samples.":[17,40,142],"For":[18],"example,":[19],"perturb":[21],"data":[22,141],"with":[23,33,47,65,74,90,155,164],"noise,":[24,148],"noise":[26,159],"is":[27],"sampled":[28],"from":[29],"Normal":[31],"distribution":[32],"fixed":[35],"standard":[36],"deviation,":[37],"for":[38,124],"We":[41,190,211],"hypothesize":[42],"that":[43],"hard":[45,72,83,150],"sample":[46,73,89,151,163],"high":[48,166],"loss":[50,93,138],"already":[51],"provides":[52],"strong":[53,76],"signal":[55],"update":[57],"model":[59],"parameters":[60,133,187],"and":[61,160,183,201,204],"should":[62,94],"be":[63,95,153],"perturbed":[64,96,154],"mild":[66],"or":[67],"no":[68],"augmentation.":[69],"Perturbing":[70],"may":[78],"also":[79],"make":[80],"it":[81],"too":[82],"learn":[85,120],"from.":[86],"Furthermore,":[87,169],"low":[91,157],"by":[97,188],"stronger":[99],"provide":[102],"more":[103],"robustness":[104],"variety":[107],"conditions.":[109],"To":[110],"formalize":[111],"these":[112],"intuitions,":[113],"we":[114],"propose":[115],"novel":[117,205],"method":[118,172,193],"Sample-Adaptive":[122],"Policy":[123],"Augmentation":[125],"\u2013":[126],"SapAugment.":[127],"Our":[128],"policy":[129,180],"adapts":[130],"based":[134],"on":[135,194,224],"In":[143],"example":[145],"Gaussian":[147],"will":[152],"variance":[158,167],"an":[161,195],"easy":[162],"noise.":[168],"proposed":[171,209],"combines":[173],"multiple":[174],"into":[177],"methodical":[179],"learning":[181],"framework":[182],"obviates":[184],"hand-crafting":[185],"trial-and-error.":[189],"our":[192],"automatic":[196],"speech":[197,230],"recognition":[198],"(ASR)":[199],"task,":[200],"combine":[202],"existing":[203],"augmentations":[206],"using":[207],"framework.":[210],"show":[212],"substantial":[213],"improvement,":[214],"up":[215],"21%":[217],"relative":[218],"reduction":[219],"in":[220],"word":[221],"error":[222],"rate":[223],"LibriSpeech":[225],"dataset,":[226],"over":[227],"state-of-the-art":[229],"method.":[232]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":5},{"year":2022,"cited_by_count":4},{"year":2021,"cited_by_count":3}],"updated_date":"2026-04-28T14:05:53.105641","created_date":"2025-10-10T00:00:00"}