{"id":"https://openalex.org/W3010802709","doi":"https://doi.org/10.48550/arxiv.2003.08773","title":"Do CNNs Encode Data Augmentations?","display_name":"Do CNNs Encode Data Augmentations?","publication_year":2020,"publication_date":"2020-02-29","ids":{"openalex":"https://openalex.org/W3010802709","doi":"https://doi.org/10.48550/arxiv.2003.08773","mag":"3010802709"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2003.08773","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2003.08773","pdf_url":"https://arxiv.org/pdf/2003.08773","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2003.08773","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5000435983","display_name":"Eddie Yan","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yan, Eddie","raw_affiliation_strings":["Allen School University of Washington Seattle , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Allen School University of Washington Seattle , USA","institution_ids":["https://openalex.org/I201448701"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5033809109","display_name":"Yanping Huang","orcid":"https://orcid.org/0000-0002-4430-3965","observed_orcids":["https://orcid.org/0000-0002-4430-3965"]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Huang, Yanping","raw_affiliation_strings":["Google Brain Mountain View , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Brain Mountain View , USA","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.913100004196167,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.913100004196167,"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"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.0617000013589859,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.004900000058114529,"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/data-augmentation","display_name":"data augmentation","score":0.9710000157356262},{"id":"https://openalex.org/keywords/convolutional-neural-networks","display_name":"convolutional neural networks","score":0.7900000214576721},{"id":"https://openalex.org/keywords/feature-encoding","display_name":"feature encoding","score":0.5120000243186951},{"id":"https://openalex.org/keywords/pre-trained-vision-models","display_name":"pre-trained vision models","score":0.7799999713897705},{"id":"https://openalex.org/keywords/robustness","display_name":"robustness","score":0.640999972820282},{"id":"https://openalex.org/keywords/color-conversion","display_name":"color conversion","score":0.7089999914169312},{"id":"https://openalex.org/keywords/brightness-and-contrast","display_name":"brightness and contrast","score":0.4650000035762787}],"concepts":[{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.8964048624038696},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6964811086654663},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6827061176300049},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6157281994819641},{"id":"https://openalex.org/C126537357","wikidata":"https://www.wikidata.org/wiki/Q372948","display_name":"Hue","level":2,"score":0.5757554769515991},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.465758353471756},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.4414580762386322},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4270970821380615},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"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/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2003.08773","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2003.08773","pdf_url":"https://arxiv.org/pdf/2003.08773","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2003.08773","doi":"https://doi.org/10.48550/arxiv.2003.08773","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2003.08773","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2003.08773","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2003.08773","pdf_url":"https://arxiv.org/pdf/2003.08773","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3010802709.pdf","grobid_xml":"https://content.openalex.org/works/W3010802709.grobid-xml"},"referenced_works_count":42,"referenced_works":["https://openalex.org/W252252322","https://openalex.org/W1800922288","https://openalex.org/W1849277567","https://openalex.org/W1932198206","https://openalex.org/W2108598243","https://openalex.org/W2141125852","https://openalex.org/W2142697503","https://openalex.org/W2158485497","https://openalex.org/W2163605009","https://openalex.org/W2169393322","https://openalex.org/W2170413022","https://openalex.org/W2194775991","https://openalex.org/W2279221249","https://openalex.org/W2295107390","https://openalex.org/W2439568532","https://openalex.org/W2526969612","https://openalex.org/W2553303224","https://openalex.org/W2746314669","https://openalex.org/W2765407302","https://openalex.org/W2785397276","https://openalex.org/W2891364477","https://openalex.org/W2913314773","https://openalex.org/W2943865428","https://openalex.org/W2946948417","https://openalex.org/W2949736877","https://openalex.org/W2950220847","https://openalex.org/W2962858109","https://openalex.org/W2963207607","https://openalex.org/W2963239066","https://openalex.org/W2963459876","https://openalex.org/W2964193438","https://openalex.org/W2965658867","https://openalex.org/W2971315489","https://openalex.org/W2992308087","https://openalex.org/W2995360807","https://openalex.org/W2995426144","https://openalex.org/W3003926725","https://openalex.org/W3035160371","https://openalex.org/W3035743198","https://openalex.org/W3104962541","https://openalex.org/W3106271444","https://openalex.org/W3161375874"],"related_works":["https://openalex.org/W4320518079","https://openalex.org/W2468279273","https://openalex.org/W2039822179","https://openalex.org/W4386771591","https://openalex.org/W2972873516","https://openalex.org/W2354198838","https://openalex.org/W1989130879","https://openalex.org/W2103419012","https://openalex.org/W2405125474","https://openalex.org/W2074864407"],"abstract_inverted_index":{"Data":[0],"augmentations":[1],"are":[2,45,113],"important":[3],"ingredients":[4],"in":[5,14,56,115,127],"the":[6,46,116,123],"recipe":[7],"for":[8],"training":[9],"robust":[10],"neural":[11,22,43,80,100],"networks,":[12],"especially":[13],"computer":[15],"vision.":[16],"A":[17],"fundamental":[18],"question":[19],"is":[20],"whether":[21],"network":[23,81],"features":[24,55,82,103,112],"encode":[25,102],"data":[26,86],"augmentation":[27,50,70,87,106,124],"transformations.":[28,51],"To":[29],"answer":[30],"this":[31],"question,":[32],"we":[33,108],"introduce":[34],"a":[35],"systematic":[36],"approach":[37,53],"to":[38,64,105],"investigate":[39],"which":[40],"layers":[41,118],"of":[42,49,119],"networks":[44,101],"most":[47],"predictive":[48],"Our":[52],"uses":[54],"pre-trained":[57],"vision":[58],"models":[59],"with":[60,94],"minimal":[61],"additional":[62],"processing":[63],"predict":[65,85,91],"common":[66],"properties":[67],"transformed":[68],"by":[69],"(scale,":[71],"aspect":[72],"ratio,":[73],"hue,":[74],"saturation,":[75],"contrast,":[76],"and":[77],"brightness).":[78],"Surprisingly,":[79],"not":[83],"only":[84],"transformations,":[88,107],"but":[89],"they":[90],"many":[92],"transformations":[93],"high":[95],"accuracy.":[96],"After":[97],"validating":[98],"that":[99,110],"corresponding":[104],"show":[109],"these":[111],"encoded":[114],"early":[117],"modern":[120],"CNNs,":[121],"though":[122],"signal":[125],"fades":[126],"deeper":[128],"layers.":[129]},"counts_by_year":[{"year":2020,"cited_by_count":1}],"updated_date":"2026-10-16T09:59:22.235173","created_date":"2025-10-10T00:00:00"}