{"id":"https://openalex.org/W4392309098","doi":"https://doi.org/10.1109/tpami.2024.3370716","title":"Randomness Regularization With Simple Consistency Training for Neural Networks","display_name":"Randomness Regularization With Simple Consistency Training for Neural Networks","publication_year":2024,"publication_date":"2024-02-29","ids":{"openalex":"https://openalex.org/W4392309098","doi":"https://doi.org/10.1109/tpami.2024.3370716","pmid":"https://pubmed.ncbi.nlm.nih.gov/38421846"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2024.3370716","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2024.3370716","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","pubmed"],"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/A5100657514","display_name":"Juntao Li","orcid":"https://orcid.org/0000-0002-6286-7529"},"institutions":[{"id":"https://openalex.org/I3923682","display_name":"Soochow University","ror":"https://ror.org/05t8y2r12","country_code":"CN","type":"education","lineage":["https://openalex.org/I3923682"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Juntao Li","raw_affiliation_strings":["Soochow University, Suzhou, China"],"affiliations":[{"raw_affiliation_string":"Soochow University, Suzhou, China","institution_ids":["https://openalex.org/I3923682"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102023923","display_name":"Xiaobo Liang","orcid":"https://orcid.org/0009-0001-1550-2877"},"institutions":[{"id":"https://openalex.org/I3923682","display_name":"Soochow University","ror":"https://ror.org/05t8y2r12","country_code":"CN","type":"education","lineage":["https://openalex.org/I3923682"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaobo Liang","raw_affiliation_strings":["Soochow University, Suzhou, China"],"affiliations":[{"raw_affiliation_string":"Soochow University, Suzhou, China","institution_ids":["https://openalex.org/I3923682"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102750692","display_name":"Lijun Wu","orcid":"https://orcid.org/0000-0002-3530-590X"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lijun Wu","raw_affiliation_strings":["Microsoft Research, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Microsoft Research, Beijing, China","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100417892","display_name":"Yue Wang","orcid":"https://orcid.org/0009-0004-7050-9811"},"institutions":[{"id":"https://openalex.org/I3923682","display_name":"Soochow University","ror":"https://ror.org/05t8y2r12","country_code":"CN","type":"education","lineage":["https://openalex.org/I3923682"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yue Wang","raw_affiliation_strings":["Soochow University, Suzhou, China"],"affiliations":[{"raw_affiliation_string":"Soochow University, Suzhou, China","institution_ids":["https://openalex.org/I3923682"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5044802273","display_name":"Qi Meng","orcid":"https://orcid.org/0000-0002-3103-1999"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qi Meng","raw_affiliation_strings":["Microsoft Research, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Microsoft Research, Beijing, China","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020025718","display_name":"Tao Qin","orcid":"https://orcid.org/0000-0002-9095-0776"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tao Qin","raw_affiliation_strings":["Microsoft Research, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Microsoft Research, Beijing, China","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100402911","display_name":"Min Zhang","orcid":"https://orcid.org/0000-0002-3895-5510"},"institutions":[{"id":"https://openalex.org/I3923682","display_name":"Soochow University","ror":"https://ror.org/05t8y2r12","country_code":"CN","type":"education","lineage":["https://openalex.org/I3923682"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Min Zhang","raw_affiliation_strings":["Soochow University, Suzhou, China"],"affiliations":[{"raw_affiliation_string":"Soochow University, Suzhou, China","institution_ids":["https://openalex.org/I3923682"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101884287","display_name":"Tie\u2010Yan Liu","orcid":"https://orcid.org/0000-0002-0476-8020"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tie-Yan Liu","raw_affiliation_strings":["Microsoft Research, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Microsoft Research, Beijing, China","institution_ids":["https://openalex.org/I4210113369"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5100657514"],"corresponding_institution_ids":["https://openalex.org/I3923682"],"apc_list":null,"apc_paid":null,"fwci":2.0851,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.88011223,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":98},"biblio":{"volume":"46","issue":"8","first_page":"5763","last_page":"5778"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9976999759674072,"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/T10320","display_name":"Neural Networks and Applications","score":0.9976999759674072,"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":0.9976000189781189,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9972000122070312,"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":0.7047320008277893},{"id":"https://openalex.org/keywords/randomness","display_name":"Randomness","score":0.7004559636116028},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6884252429008484},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.669826865196228},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6193471550941467},{"id":"https://openalex.org/keywords/simple","display_name":"Simple (philosophy)","score":0.5678020119667053},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.5150930285453796},{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.4688274562358856},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.4463363289833069},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.44437921047210693},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.19526642560958862},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.11129829287528992}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7047320008277893},{"id":"https://openalex.org/C125112378","wikidata":"https://www.wikidata.org/wiki/Q176640","display_name":"Randomness","level":2,"score":0.7004559636116028},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6884252429008484},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.669826865196228},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6193471550941467},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.5678020119667053},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5150930285453796},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.4688274562358856},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4463363289833069},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.44437921047210693},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.19526642560958862},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.11129829287528992},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","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},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tpami.2024.3370716","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2024.3370716","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},{"id":"pmid:38421846","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/38421846","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE transactions on pattern analysis and machine intelligence","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2027526133","display_name":null,"funder_award_id":"BK20220488","funder_id":"https://openalex.org/F4320322769","funder_display_name":"Natural Science Foundation of Jiangsu Province"},{"id":"https://openalex.org/G6441875606","display_name":null,"funder_award_id":"62206194","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320322769","display_name":"Natural Science Foundation of Jiangsu Province","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":133,"referenced_works":["https://openalex.org/W1034159276","https://openalex.org/W1519506695","https://openalex.org/W1522301498","https://openalex.org/W1677182931","https://openalex.org/W1821462560","https://openalex.org/W1904365287","https://openalex.org/W2095705004","https://openalex.org/W2108598243","https://openalex.org/W2130158090","https://openalex.org/W2133564696","https://openalex.org/W2160204597","https://openalex.org/W2183341477","https://openalex.org/W2250473257","https://openalex.org/W2479750863","https://openalex.org/W2604847698","https://openalex.org/W2620998106","https://openalex.org/W2746314669","https://openalex.org/W2759764766","https://openalex.org/W2896457183","https://openalex.org/W2911484737","https://openalex.org/W2923014074","https://openalex.org/W2933138175","https://openalex.org/W2949615363","https://openalex.org/W2962784628","https://openalex.org/W2963532001","https://openalex.org/W2963654130","https://openalex.org/W2963744496","https://openalex.org/W2963807318","https://openalex.org/W2964093309","https://openalex.org/W2965373594","https://openalex.org/W2979826702","https://openalex.org/W2983128379","https://openalex.org/W2987861506","https://openalex.org/W2989571009","https://openalex.org/W2993587506","https://openalex.org/W3006051380","https://openalex.org/W3034999214","https://openalex.org/W3066373881","https://openalex.org/W3089659770","https://openalex.org/W3097217077","https://openalex.org/W3098985395","https://openalex.org/W3100439847","https://openalex.org/W3113303810","https://openalex.org/W3118608800","https://openalex.org/W3138516171","https://openalex.org/W3153675281","https://openalex.org/W3155713635","https://openalex.org/W3172096628","https://openalex.org/W3173982024","https://openalex.org/W3176828726","https://openalex.org/W3177318507","https://openalex.org/W3197681757","https://openalex.org/W4205991051","https://openalex.org/W4214488463","https://openalex.org/W4250482878","https://openalex.org/W4287391717","https://openalex.org/W4287692509","https://openalex.org/W4288364646","https://openalex.org/W4292779060","https://openalex.org/W4295838474","https://openalex.org/W4298422451","https://openalex.org/W4312544276","https://openalex.org/W4385245566","https://openalex.org/W4386075867","https://openalex.org/W4386187806","https://openalex.org/W6600213771","https://openalex.org/W6601425427","https://openalex.org/W6631160641","https://openalex.org/W6631190155","https://openalex.org/W6632455782","https://openalex.org/W6638523607","https://openalex.org/W6640036494","https://openalex.org/W6674330103","https://openalex.org/W6676984168","https://openalex.org/W6677604277","https://openalex.org/W6679434410","https://openalex.org/W6679955943","https://openalex.org/W6680910308","https://openalex.org/W6681151457","https://openalex.org/W6684191040","https://openalex.org/W6685053522","https://openalex.org/W6688167117","https://openalex.org/W6691459498","https://openalex.org/W6695676441","https://openalex.org/W6697814554","https://openalex.org/W6727099177","https://openalex.org/W6727785063","https://openalex.org/W6732814185","https://openalex.org/W6738735913","https://openalex.org/W6742632731","https://openalex.org/W6743428213","https://openalex.org/W6745831770","https://openalex.org/W6751751081","https://openalex.org/W6754905691","https://openalex.org/W6754929296","https://openalex.org/W6756358366","https://openalex.org/W6758017961","https://openalex.org/W6758684365","https://openalex.org/W6759579507","https://openalex.org/W6762269264","https://openalex.org/W6763701032","https://openalex.org/W6764043288","https://openalex.org/W6764679822","https://openalex.org/W6766279884","https://openalex.org/W6766673545","https://openalex.org/W6768080748","https://openalex.org/W6770855403","https://openalex.org/W6771626834","https://openalex.org/W6771713106","https://openalex.org/W6771915120","https://openalex.org/W6771917389","https://openalex.org/W6772452955","https://openalex.org/W6773837390","https://openalex.org/W6774085601","https://openalex.org/W6776488958","https://openalex.org/W6778883912","https://openalex.org/W6780226713","https://openalex.org/W6781421678","https://openalex.org/W6782699318","https://openalex.org/W6783432658","https://openalex.org/W6783973398","https://openalex.org/W6784286987","https://openalex.org/W6784333009","https://openalex.org/W6787473312","https://openalex.org/W6787972765","https://openalex.org/W6788280241","https://openalex.org/W6788811087","https://openalex.org/W6789954222","https://openalex.org/W6794640596","https://openalex.org/W6796581206","https://openalex.org/W6797110910","https://openalex.org/W6797386277","https://openalex.org/W6802298853"],"related_works":["https://openalex.org/W3034924094","https://openalex.org/W3094954546","https://openalex.org/W1488708774","https://openalex.org/W1982811510","https://openalex.org/W4391100477","https://openalex.org/W2402189625","https://openalex.org/W4327779705","https://openalex.org/W4310560702","https://openalex.org/W1513698804","https://openalex.org/W2029712093"],"abstract_inverted_index":{"Randomness":[0],"is":[1,156],"widely":[2],"introduced":[3,47],"in":[4,23,32,151],"neural":[5,38,163,170],"network":[6],"training":[7,35,56,67,105],"to":[8,69,86],"simplify":[9],"model":[10,28,137,190],"optimization":[11],"or":[12],"avoid":[13],"the":[14,34,45,92,114,119,122,128,131,135,187],"over-fitting":[15],"problem.":[16],"Among":[17],"them,":[18],"dropout":[19],"and":[20,42,57,126,138,168,172,178,200,217],"its":[21],"variations":[22],"different":[24,160,173],"aspects":[25],"(e.g.,":[26],"data,":[27],"structure)":[29],"are":[30],"prevalent":[31],"regularizing":[33],"of":[36,84,162,221],"deep":[37,145],"networks.":[39],"Though":[40],"effective":[41,158],"performing":[43],"well,":[44],"randomness":[46,85,102],"by":[48,81,100,117,134],"these":[49],"dropout-based":[50,101],"methods":[51],"causes":[52],"nonnegligible":[53],"inconsistency":[54,116,120],"between":[55,95,130],"inference.":[58],"In":[59,180],"this":[60],"paper,":[61],"we":[62],"introduce":[63],"a":[64],"simple":[65],"consistency":[66],"strategy":[68],"regularize":[70],"such":[71],"randomness,":[72],"namely":[73],"R-Drop,":[74],"which":[75],"forces":[76],"two":[77,96],"output":[78,97],"distributions":[79,98],"sampled":[80,123],"each":[82,104],"type":[83],"be":[87],"consistent.":[88],"Specifically,":[89],"R-Drop":[90,111,155],"minimizes":[91],"bidirectional":[93],"KL-divergence":[94],"produced":[99],"for":[103,159],"sample.":[106],"Theoretical":[107],"analysis":[108],"reveals":[109],"that":[110,154],"can":[112],"reduce":[113],"above":[115],"reducing":[118],"among":[121],"sub":[124,139],"structures":[125],"bridging":[127],"gap":[129],"loss":[132],"calculated":[133],"full":[136],"structures.":[140],"Experiments":[141],"on":[142,191],"7":[143],"widely-used":[144],"learning":[146,174],"tasks":[147],"(":[148,197,206],"23":[149],"datasets":[150],"total)":[152],"demonstrate":[153],"universally":[157],"types":[161],"networks":[164],"(i.e.,":[165],"feed-forward,":[166],"recurrent,":[167],"graph":[169],"networks)":[171],"paradigms":[175],"(supervised,":[176],"parameter-efficient,":[177],"semi-supervised).":[179],"particular,":[181],"it":[182],"achieves":[183],"state-of-the-art":[184],"performances":[185],"with":[186,213],"vanilla":[188],"Transformer":[189,222],"WMT14":[192,201],"English":[193,202],"\u2192":[194,203],"German":[195],"translation":[196,205],"30.91":[198],"BLEU)":[199],"French":[204],"43.95":[207],"BLEU),":[208],"even":[209],"surpassing":[210],"models":[211],"trained":[212],"extra":[214],"large-scale":[215],"data":[216],"expert-designed":[218],"advanced":[219],"variants":[220],"models.":[223]},"counts_by_year":[{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":1}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}