{"id":"https://openalex.org/W4372260407","doi":"https://doi.org/10.1109/icassp49357.2023.10095552","title":"Wordreg: Mitigating the Gap between Training and Inference with Worst-Case Drop Regularization","display_name":"Wordreg: Mitigating the Gap between Training and Inference with Worst-Case Drop Regularization","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4372260407","doi":"https://doi.org/10.1109/icassp49357.2023.10095552"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49357.2023.10095552","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/icassp49357.2023.10095552","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5087672753","display_name":"Jun Xia","orcid":"https://orcid.org/0000-0002-7993-0803"},"institutions":[{"id":"https://openalex.org/I3133055985","display_name":"Westlake University","ror":"https://ror.org/05hfa4n20","country_code":"CN","type":"education","lineage":["https://openalex.org/I3133055985"]},{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jun Xia","raw_affiliation_strings":["Zhejiang University,Hangzhou,China,310058","School of Engineering, Westlake University, Hangzhou, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University,Hangzhou,China,310058","institution_ids":["https://openalex.org/I76130692"]},{"raw_affiliation_string":"School of Engineering, Westlake University, Hangzhou, China","institution_ids":["https://openalex.org/I3133055985"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100400458","display_name":"Ge Wang","orcid":"https://orcid.org/0000-0002-2656-7705"},"institutions":[{"id":"https://openalex.org/I3133055985","display_name":"Westlake University","ror":"https://ror.org/05hfa4n20","country_code":"CN","type":"education","lineage":["https://openalex.org/I3133055985"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ge Wang","raw_affiliation_strings":["Westlake University,School of Engineering,Hangzhou,China,310030"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Westlake University,School of Engineering,Hangzhou,China,310030","institution_ids":["https://openalex.org/I3133055985"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012276767","display_name":"Bozhen Hu","orcid":"https://orcid.org/0000-0002-3428-0114"},"institutions":[{"id":"https://openalex.org/I3133055985","display_name":"Westlake University","ror":"https://ror.org/05hfa4n20","country_code":"CN","type":"education","lineage":["https://openalex.org/I3133055985"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bozhen Hu","raw_affiliation_strings":["Westlake University,School of Engineering,Hangzhou,China,310030"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Westlake University,School of Engineering,Hangzhou,China,310030","institution_ids":["https://openalex.org/I3133055985"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006542157","display_name":"Cheng Tan","orcid":"https://orcid.org/0000-0002-8639-923X"},"institutions":[{"id":"https://openalex.org/I3133055985","display_name":"Westlake University","ror":"https://ror.org/05hfa4n20","country_code":"CN","type":"education","lineage":["https://openalex.org/I3133055985"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Cheng Tan","raw_affiliation_strings":["Westlake University,School of Engineering,Hangzhou,China,310030"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Westlake University,School of Engineering,Hangzhou,China,310030","institution_ids":["https://openalex.org/I3133055985"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101808238","display_name":"Jiangbin Zheng","orcid":"https://orcid.org/0000-0003-3305-0103"},"institutions":[{"id":"https://openalex.org/I3133055985","display_name":"Westlake University","ror":"https://ror.org/05hfa4n20","country_code":"CN","type":"education","lineage":["https://openalex.org/I3133055985"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiangbin Zheng","raw_affiliation_strings":["Westlake University,School of Engineering,Hangzhou,China,310030"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Westlake University,School of Engineering,Hangzhou,China,310030","institution_ids":["https://openalex.org/I3133055985"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042875688","display_name":"Yongjie Xu","orcid":"https://orcid.org/0000-0002-2660-6575"},"institutions":[{"id":"https://openalex.org/I3133055985","display_name":"Westlake University","ror":"https://ror.org/05hfa4n20","country_code":"CN","type":"education","lineage":["https://openalex.org/I3133055985"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yongjie Xu","raw_affiliation_strings":["Westlake University,School of Engineering,Hangzhou,China,310030"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Westlake University,School of Engineering,Hangzhou,China,310030","institution_ids":["https://openalex.org/I3133055985"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5082786719","display_name":"Stan Z. Li","orcid":"https://orcid.org/0000-0002-2961-8096"},"institutions":[{"id":"https://openalex.org/I3133055985","display_name":"Westlake University","ror":"https://ror.org/05hfa4n20","country_code":"CN","type":"education","lineage":["https://openalex.org/I3133055985"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Stan Z. Li","raw_affiliation_strings":["Westlake University,School of Engineering,Hangzhou,China,310030"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Westlake University,School of Engineering,Hangzhou,China,310030","institution_ids":["https://openalex.org/I3133055985"]}]}],"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":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9995999932289124,"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":0.9993000030517578,"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/inference","display_name":"Inference","score":0.7678707838058472},{"id":"https://openalex.org/keywords/dropout","display_name":"Dropout (neural networks)","score":0.7468003034591675},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.7339763641357422},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6468650698661804},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.6303896307945251},{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.5541403889656067},{"id":"https://openalex.org/keywords/divergence","display_name":"Divergence (linguistics)","score":0.4917011857032776},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.4563501477241516},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4499172568321228},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4360136091709137},{"id":"https://openalex.org/keywords/approximate-inference","display_name":"Approximate inference","score":0.4312084913253784},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4200612008571625}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7678707838058472},{"id":"https://openalex.org/C2776145597","wikidata":"https://www.wikidata.org/wiki/Q25339462","display_name":"Dropout (neural networks)","level":2,"score":0.7468003034591675},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.7339763641357422},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6468650698661804},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.6303896307945251},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.5541403889656067},{"id":"https://openalex.org/C207390915","wikidata":"https://www.wikidata.org/wiki/Q1230525","display_name":"Divergence (linguistics)","level":2,"score":0.4917011857032776},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.4563501477241516},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4499172568321228},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4360136091709137},{"id":"https://openalex.org/C2777472644","wikidata":"https://www.wikidata.org/wiki/Q16968992","display_name":"Approximate inference","level":3,"score":0.4312084913253784},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4200612008571625},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"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/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49357.2023.10095552","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/icassp49357.2023.10095552","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320318567","display_name":"Westlake University","ror":null},{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":38,"referenced_works":["https://openalex.org/W4919037","https://openalex.org/W1595903022","https://openalex.org/W2095705004","https://openalex.org/W2135762244","https://openalex.org/W2194775991","https://openalex.org/W2519887557","https://openalex.org/W2896457183","https://openalex.org/W2923014074","https://openalex.org/W2962711740","https://openalex.org/W2963654130","https://openalex.org/W2964015378","https://openalex.org/W2965373594","https://openalex.org/W2970597249","https://openalex.org/W3042313988","https://openalex.org/W3094502228","https://openalex.org/W3121904315","https://openalex.org/W3190486510","https://openalex.org/W3205655893","https://openalex.org/W4221023051","https://openalex.org/W4225310585","https://openalex.org/W4285209827","https://openalex.org/W4287824654","https://openalex.org/W4310557809","https://openalex.org/W4365455552","https://openalex.org/W6600213771","https://openalex.org/W6674330103","https://openalex.org/W6727785063","https://openalex.org/W6754929296","https://openalex.org/W6755207826","https://openalex.org/W6763701032","https://openalex.org/W6764043288","https://openalex.org/W6766673545","https://openalex.org/W6771917389","https://openalex.org/W6780444311","https://openalex.org/W6784333009","https://openalex.org/W6788752698","https://openalex.org/W6847339387","https://openalex.org/W6851073391"],"related_works":["https://openalex.org/W3082178636","https://openalex.org/W2782041652","https://openalex.org/W2612657834","https://openalex.org/W2392157706","https://openalex.org/W2599192953","https://openalex.org/W2952088488","https://openalex.org/W1521968289","https://openalex.org/W4225691210","https://openalex.org/W4220659530","https://openalex.org/W1937880684"],"abstract_inverted_index":{"Dropout":[0],"has":[1],"emerged":[2],"as":[3],"one":[4],"of":[5,60,96],"the":[6,19,30,55,58,70,75,78,91,94,97,134],"most":[7],"frequently":[8],"used":[9],"techniques":[10],"for":[11],"training":[12,26,137],"deep":[13],"neural":[14],"networks":[15],"(DNNs).":[16],"Although":[17],"effective,":[18],"sampled":[20,63],"sub-model":[21],"by":[22,64,73],"random":[23,85],"dropout":[24,52,72],"during":[25,35],"is":[27],"inconsistent":[28],"with":[29,80,83,100],"full":[31],"model":[32],"(without":[33],"dropout)":[34],"inference.":[36,139],"To":[37],"mitigate":[38],"this":[39],"undesirable":[40],"gap,":[41],"we":[42,128],"propose":[43],"WordReg,":[44],"a":[45],"simple":[46],"yet":[47],"effective":[48],"regularization":[49],"built":[50],"on":[51,105],"that":[53,111,130],"enforces":[54],"consistency":[56],"between":[57,77,93,136],"outputs":[59,79,95],"different":[61,84],"sub-models":[62,82,99],"dropout.":[65],"Specifically,":[66],"WordReg":[67,112,131],"first":[68],"obtains":[69],"worst-case":[71],"maximizing":[74],"divergence":[76],"two":[81,98],"dropouts.":[86],"And":[87],"then,":[88],"it":[89],"encourages":[90],"agreements":[92],"worstcase":[101],"divergence.":[102],"Extensive":[103],"experiments":[104],"diverse":[106],"DNNs":[107],"and":[108,116,122,138],"tasks":[109],"reveal":[110],"can":[113,132],"achieve":[114],"notable":[115],"consistent":[117],"improvements":[118],"over":[119],"non-regularized":[120],"models":[121],"yields":[123],"some":[124],"state-of-the-art":[125],"results.":[126],"Theoretically,":[127],"verify":[129],"reduce":[133],"gap":[135]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}