{"id":"https://openalex.org/W2954003412","doi":"https://doi.org/10.1109/ijcnn.2019.8851850","title":"Improve L2-normalized Softmax with Exponential Moving Average","display_name":"Improve L2-normalized Softmax with Exponential Moving Average","publication_year":2019,"publication_date":"2019-07-01","ids":{"openalex":"https://openalex.org/W2954003412","doi":"https://doi.org/10.1109/ijcnn.2019.8851850","mag":"2954003412"},"language":"en","primary_location":{"id":"doi:10.1109/ijcnn.2019.8851850","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn.2019.8851850","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2019 International Joint Conference on Neural Networks (IJCNN)","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/A5044236581","display_name":"Xuefei Zhe","orcid":"https://orcid.org/0000-0002-5005-7166"},"institutions":[{"id":"https://openalex.org/I168719708","display_name":"City University of Hong Kong","ror":"https://ror.org/03q8dnn23","country_code":"HK","type":"education","lineage":["https://openalex.org/I168719708"]}],"countries":["HK"],"is_corresponding":true,"raw_author_name":"Xuefei Zhe","raw_affiliation_strings":["Department of Electronic Engineering, City University of Hong Kong, Hong Kong, China"],"affiliations":[{"raw_affiliation_string":"Department of Electronic Engineering, City University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I168719708"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030129372","display_name":"Le Ou-Yang","orcid":"https://orcid.org/0000-0003-4007-4568"},"institutions":[{"id":"https://openalex.org/I180726961","display_name":"Shenzhen University","ror":"https://ror.org/01vy4gh70","country_code":"CN","type":"education","lineage":["https://openalex.org/I180726961"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Le Ou-Yang","raw_affiliation_strings":["Guangdong key Laboratory of intelligent information Processing Shenzhen key Laboratory of Media Security, Shenzhen University, Guangdong, China"],"affiliations":[{"raw_affiliation_string":"Guangdong key Laboratory of intelligent information Processing Shenzhen key Laboratory of Media Security, Shenzhen University, Guangdong, China","institution_ids":["https://openalex.org/I180726961"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100644375","display_name":"Hong Yan","orcid":"https://orcid.org/0000-0001-9661-3095"},"institutions":[{"id":"https://openalex.org/I168719708","display_name":"City University of Hong Kong","ror":"https://ror.org/03q8dnn23","country_code":"HK","type":"education","lineage":["https://openalex.org/I168719708"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Hong Yan","raw_affiliation_strings":["Department of Electronic Engineering, City University of Hong Kong, Hong Kong, China"],"affiliations":[{"raw_affiliation_string":"Department of Electronic Engineering, City University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I168719708"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5044236581"],"corresponding_institution_ids":["https://openalex.org/I168719708"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.06227391,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.9994999766349792,"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/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.9994999766349792,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9984999895095825,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.9979000091552734,"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/softmax-function","display_name":"Softmax function","score":0.9925912022590637},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6753417253494263},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.6426092386245728},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6118971109390259},{"id":"https://openalex.org/keywords/exponential-function","display_name":"Exponential function","score":0.5464748740196228},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.4668505787849426},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.4442558288574219},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.44204509258270264},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.4407529830932617},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.42345455288887024},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.35663264989852905},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.35483184456825256},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.21555644273757935}],"concepts":[{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.9925912022590637},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6753417253494263},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.6426092386245728},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6118971109390259},{"id":"https://openalex.org/C151376022","wikidata":"https://www.wikidata.org/wiki/Q168698","display_name":"Exponential function","level":2,"score":0.5464748740196228},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.4668505787849426},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.4442558288574219},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.44204509258270264},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.4407529830932617},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.42345455288887024},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.35663264989852905},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35483184456825256},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.21555644273757935},{"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/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ijcnn.2019.8851850","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn.2019.8851850","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2019 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":52,"referenced_works":["https://openalex.org/W6908809","https://openalex.org/W1522301498","https://openalex.org/W1533861849","https://openalex.org/W1836465849","https://openalex.org/W1977295328","https://openalex.org/W2096733369","https://openalex.org/W2108598243","https://openalex.org/W2112796928","https://openalex.org/W2124509324","https://openalex.org/W2138011018","https://openalex.org/W2163605009","https://openalex.org/W2186615578","https://openalex.org/W2194775991","https://openalex.org/W2520774990","https://openalex.org/W2533598788","https://openalex.org/W2549139847","https://openalex.org/W2549143551","https://openalex.org/W2549714022","https://openalex.org/W2555897561","https://openalex.org/W2600537992","https://openalex.org/W2609575245","https://openalex.org/W2613997951","https://openalex.org/W2625715238","https://openalex.org/W2750672897","https://openalex.org/W2798869704","https://openalex.org/W2897275116","https://openalex.org/W2962798326","https://openalex.org/W2962887033","https://openalex.org/W2962898354","https://openalex.org/W2963026686","https://openalex.org/W2963446712","https://openalex.org/W2963466847","https://openalex.org/W2964121744","https://openalex.org/W2969985801","https://openalex.org/W3099206234","https://openalex.org/W3101227480","https://openalex.org/W4293478066","https://openalex.org/W6600284362","https://openalex.org/W6631190155","https://openalex.org/W6631943919","https://openalex.org/W6638667902","https://openalex.org/W6684191040","https://openalex.org/W6686509673","https://openalex.org/W6693735323","https://openalex.org/W6725739302","https://openalex.org/W6730323794","https://openalex.org/W6738819434","https://openalex.org/W6743186171","https://openalex.org/W6748010250","https://openalex.org/W6748522371","https://openalex.org/W6750348571","https://openalex.org/W6842019321"],"related_works":["https://openalex.org/W2978855952","https://openalex.org/W2762456960","https://openalex.org/W3098169841","https://openalex.org/W3018925616","https://openalex.org/W2735411683","https://openalex.org/W2996955463","https://openalex.org/W3045804464","https://openalex.org/W2261232712","https://openalex.org/W2804306369","https://openalex.org/W2900628392","https://openalex.org/W2890160131","https://openalex.org/W2765897105","https://openalex.org/W2787021484","https://openalex.org/W2884347112","https://openalex.org/W3005381706","https://openalex.org/W2890311316","https://openalex.org/W3035635935","https://openalex.org/W2483754710","https://openalex.org/W2890619223","https://openalex.org/W2762473905"],"abstract_inverted_index":{"In":[0,100],"this":[1,80,101],"paper,":[2,102],"we":[3,103,126],"propose":[4,127],"an":[5,64],"effective":[6],"training":[7,120],"method":[8],"to":[9,89,91,161],"improve":[10],"the":[11,29,34,49,72,76,105,109,113,118,134,140,162],"performance":[12,154],"of":[13,22,32,36,67,79,107],"L2-normalized":[14,50,114],"softmax":[15,60],"for":[16],"convolutional":[17],"neural":[18],"networks.":[19],"Recent":[20],"studies":[21],"deep":[23],"learning":[24,108,130],"show":[25,146],"that":[26,63,132,147],"by":[27,56],"L2-normalizing":[28],"input":[30],"features":[31],"softmax,":[33],"accuracy":[35],"CNN":[37,119],"can":[38],"be":[39],"increased.":[40],"Several":[41],"works":[42],"proposed":[43],"novel":[44,129],"loss":[45],"functions":[46],"based":[47],"on":[48],"softmax.":[51,115],"A":[52],"common":[53],"property":[54],"shared":[55],"these":[57,93],"modified":[58],"normalized":[59],"models":[61],"is":[62,69,82],"extra":[65],"set":[66],"parameters":[68],"introduced":[70],"as":[71,122],"class":[73,94,110],"centers.":[74],"Although":[75],"physical":[77],"meaning":[78],"parameter":[81],"clear,":[83],"few":[84],"attentions":[85],"have":[86],"been":[87],"paid":[88],"how":[90],"learn":[92],"centers,":[95],"which":[96],"limits":[97],"further":[98],"improvement.":[99],"address":[104],"problem":[106],"centers":[111],"in":[112],"By":[116],"treating":[117],"process":[121],"a":[123,128,158],"time":[124],"series,":[125],"algorithm":[131],"combines":[133],"generally":[135],"used":[136],"gradient":[137],"descent":[138],"with":[139],"exponential":[141],"moving":[142],"average.":[143],"Extensive":[144],"experiments":[145],"our":[148],"model":[149],"not":[150],"only":[151],"achieves":[152],"better":[153],"but":[155],"also":[156],"has":[157],"higher":[159],"tolerance":[160],"imbalance":[163],"data.":[164]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}