{"id":"https://openalex.org/W3137609883","doi":"https://doi.org/10.1109/tpami.2021.3067100","title":"Self-Distillation: Towards Efficient and Compact Neural Networks","display_name":"Self-Distillation: Towards Efficient and Compact Neural Networks","publication_year":2021,"publication_date":"2021-01-01","ids":{"openalex":"https://openalex.org/W3137609883","doi":"https://doi.org/10.1109/tpami.2021.3067100","mag":"3137609883","pmid":"https://pubmed.ncbi.nlm.nih.gov/33735074"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2021.3067100","is_oa":true,"landing_page_url":"https://doi.org/10.1109/tpami.2021.3067100","pdf_url":"https://ieeexplore.ieee.org/ielx7/34/9812931/09381661.pdf","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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":true,"oa_status":"hybrid","oa_url":"https://ieeexplore.ieee.org/ielx7/34/9812931/09381661.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100689114","display_name":"Linfeng Zhang","orcid":"https://orcid.org/0000-0002-3341-183X"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Linfeng Zhang","raw_affiliation_strings":["Institute of Interdisciplinary Information Sciences, Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Interdisciplinary Information Sciences, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054041500","display_name":"Chenglong Bao","orcid":"https://orcid.org/0000-0002-1201-1212"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chenglong Bao","raw_affiliation_strings":["Yau Mathematical Sciences Center, Yanqi Lake Beijing Institute of Mathematical Sciences and Applications, Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Yau Mathematical Sciences Center, Yanqi Lake Beijing Institute of Mathematical Sciences and Applications, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5006570986","display_name":"Kaisheng Ma","orcid":"https://orcid.org/0000-0001-9226-3366"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Kaisheng Ma","raw_affiliation_strings":["Institute of Interdisciplinary Information Sciences, Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Interdisciplinary Information Sciences, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5100689114"],"corresponding_institution_ids":["https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":13.9824,"has_fulltext":true,"cited_by_count":211,"citation_normalized_percentile":{"value":0.99293909,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":100},"biblio":{"volume":"44","issue":"8","first_page":"1","last_page":"1"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998999834060669,"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":0.9998999834060669,"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/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.9983999729156494,"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":0.998199999332428,"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/distillation","display_name":"Distillation","score":0.8802835941314697},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.7572145462036133},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7433276176452637},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6202932596206665},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.5561641454696655},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.48874402046203613},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.48075100779533386},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.14737966656684875},{"id":"https://openalex.org/keywords/chemistry","display_name":"Chemistry","score":0.09470859169960022}],"concepts":[{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.8802835941314697},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.7572145462036133},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7433276176452637},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6202932596206665},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5561641454696655},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.48874402046203613},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.48075100779533386},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.14737966656684875},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.09470859169960022},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","level":1,"score":0.0},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0}],"mesh":[{"descriptor_ui":"D000465","descriptor_name":"Algorithms","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D000465","descriptor_name":"Algorithms","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D000465","descriptor_name":"Algorithms","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D006801","descriptor_name":"Humans","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D006801","descriptor_name":"Humans","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D006801","descriptor_name":"Humans","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D016571","descriptor_name":"Neural Networks, Computer","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D016571","descriptor_name":"Neural Networks, Computer","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D016571","descriptor_name":"Neural Networks, Computer","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true}],"locations_count":2,"locations":[{"id":"doi:10.1109/tpami.2021.3067100","is_oa":true,"landing_page_url":"https://doi.org/10.1109/tpami.2021.3067100","pdf_url":"https://ieeexplore.ieee.org/ielx7/34/9812931/09381661.pdf","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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:33735074","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/33735074","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":{"id":"doi:10.1109/tpami.2021.3067100","is_oa":true,"landing_page_url":"https://doi.org/10.1109/tpami.2021.3067100","pdf_url":"https://ieeexplore.ieee.org/ielx7/34/9812931/09381661.pdf","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.5799999833106995,"display_name":"Quality Education"}],"awards":[{"id":"https://openalex.org/G2031701609","display_name":null,"funder_award_id":"31970972","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G5111578883","display_name":null,"funder_award_id":"11901338","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/F4320322392","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3137609883.pdf","grobid_xml":"https://content.openalex.org/works/W3137609883.grobid-xml"},"referenced_works_count":93,"referenced_works":["https://openalex.org/W753847829","https://openalex.org/W1686810756","https://openalex.org/W1901129140","https://openalex.org/W1920022804","https://openalex.org/W1996901117","https://openalex.org/W2108598243","https://openalex.org/W2146260356","https://openalex.org/W2173379916","https://openalex.org/W2194775991","https://openalex.org/W2279098554","https://openalex.org/W2294370754","https://openalex.org/W2300242332","https://openalex.org/W2549139847","https://openalex.org/W2562731582","https://openalex.org/W2604790786","https://openalex.org/W2618530766","https://openalex.org/W2620998106","https://openalex.org/W2739879705","https://openalex.org/W2752782242","https://openalex.org/W2798170643","https://openalex.org/W2802198257","https://openalex.org/W2883780447","https://openalex.org/W2884561390","https://openalex.org/W2884585870","https://openalex.org/W2886851211","https://openalex.org/W2901505625","https://openalex.org/W2952787292","https://openalex.org/W2962944050","https://openalex.org/W2963037989","https://openalex.org/W2963140444","https://openalex.org/W2963150697","https://openalex.org/W2963163009","https://openalex.org/W2963393494","https://openalex.org/W2963446712","https://openalex.org/W2964016283","https://openalex.org/W2964137095","https://openalex.org/W2975480662","https://openalex.org/W2981698279","https://openalex.org/W2981751377","https://openalex.org/W2981819252","https://openalex.org/W2982083293","https://openalex.org/W2982157312","https://openalex.org/W2982220706","https://openalex.org/W2982242214","https://openalex.org/W2984618279","https://openalex.org/W2987861506","https://openalex.org/W2990045899","https://openalex.org/W2990231018","https://openalex.org/W2990792015","https://openalex.org/W2996999409","https://openalex.org/W2997006708","https://openalex.org/W3004127093","https://openalex.org/W3009751875","https://openalex.org/W3016719260","https://openalex.org/W3034406766","https://openalex.org/W3091093673","https://openalex.org/W3104263540","https://openalex.org/W3106153666","https://openalex.org/W3118608800","https://openalex.org/W4293390340","https://openalex.org/W4295727797","https://openalex.org/W4297775537","https://openalex.org/W4375869345","https://openalex.org/W6620707391","https://openalex.org/W6637551013","https://openalex.org/W6638523607","https://openalex.org/W6639703010","https://openalex.org/W6677103964","https://openalex.org/W6678583879","https://openalex.org/W6684563725","https://openalex.org/W6684665197","https://openalex.org/W6685316698","https://openalex.org/W6691815588","https://openalex.org/W6695314431","https://openalex.org/W6726983090","https://openalex.org/W6730179637","https://openalex.org/W6734062232","https://openalex.org/W6745447533","https://openalex.org/W6745499552","https://openalex.org/W6751444130","https://openalex.org/W6751751081","https://openalex.org/W6751795773","https://openalex.org/W6751979845","https://openalex.org/W6753038380","https://openalex.org/W6755843862","https://openalex.org/W6756633790","https://openalex.org/W6757036269","https://openalex.org/W6762100871","https://openalex.org/W6763094690","https://openalex.org/W6763260857","https://openalex.org/W6765492279","https://openalex.org/W6768165065","https://openalex.org/W6779253532"],"related_works":["https://openalex.org/W2373300491","https://openalex.org/W2395294869","https://openalex.org/W2378744544","https://openalex.org/W2594301978","https://openalex.org/W2379704676","https://openalex.org/W1998810860","https://openalex.org/W4206442282","https://openalex.org/W2384505857","https://openalex.org/W2355171581","https://openalex.org/W4229439743"],"abstract_inverted_index":{"Remarkable":[0],"achievements":[1],"have":[2],"been":[3],"obtained":[4],"by":[5,24],"deep":[6],"neural":[7,18,68,129,155],"networks":[8,19,69,156],"in":[9,17,108,125,133],"the":[10,15,74,78,83,89,92,109,114,118,122,128],"last":[11],"several":[12,58],"years.":[13],"However,":[14],"breakthrough":[16],"accuracy":[20,165],"is":[21,95],"always":[22],"accompanied":[23],"explosive":[25],"growth":[26],"of":[27,37,67,91],"computation":[28],"and":[29,61,70,150,157,162,171,190],"parameters,":[30],"which":[31,137],"leads":[32,138],"to":[33,52,77,97,117,131,139],"a":[34,45,134,140],"severe":[35],"limitation":[36],"model":[38,94,111,183,192],"deployment.":[39],"In":[40],"this":[41,54],"paper,":[42],"we":[43],"propose":[44],"novel":[46],"knowledge":[47,72,85,90,106,187],"distillation":[48,86],"technique":[49],"named":[50],"self-distillation":[51,101,126,147,177],"address":[53],"problem.":[55],"Self-distillation":[56],"attaches":[57],"attention":[59],"modules":[60],"shallow":[62,119],"classifiers":[63,124],"at":[64],"different":[65],"depths":[66],"distills":[71],"from":[73,82,113],"deepest":[75],"classifier":[76],"shallower":[79],"classifiers.":[80],"Different":[81],"conventional":[84],"methods":[87],"where":[88],"teacher":[93],"transferred":[96],"another":[98],"student":[99],"model,":[100],"can":[102,178],"be":[103,179],"considered":[104],"as":[105],"transfer":[107],"same":[110],"-":[112],"deeper":[115],"layers":[116],"layers.":[120],"Moreover,":[121],"additional":[123],"allow":[127],"network":[130],"work":[132],"dynamic":[135],"manner,":[136],"much":[141],"higher":[142],"acceleration.":[143],"Experiments":[144],"demonstrate":[145],"that":[146,176],"has":[148],"consistent":[149],"significant":[151],"effectiveness":[152],"on":[153,169],"various":[154],"datasets.":[158],"On":[159],"average,":[160],"3.49":[161],"2.32":[163],"percent":[164],"boost":[166],"are":[167],"observed":[168],"CIFAR100":[170],"ImageNet.":[172],"Besides,":[173],"experiments":[174],"show":[175],"combined":[180],"with":[181],"other":[182],"compression":[184],"methods,":[185],"including":[186],"distillation,":[188],"pruning":[189],"lightweight":[191],"design.":[193]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":63},{"year":2024,"cited_by_count":69},{"year":2023,"cited_by_count":50},{"year":2022,"cited_by_count":22},{"year":2021,"cited_by_count":3}],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-10T00:00:00"}