{"id":"https://openalex.org/W3182242813","doi":"https://doi.org/10.24963/ijcai.2020/136","title":"TRP: Trained Rank Pruning for Efficient Deep Neural Networks","display_name":"TRP: Trained Rank Pruning for Efficient Deep Neural Networks","publication_year":2020,"publication_date":"2020-07-01","ids":{"openalex":"https://openalex.org/W3182242813","doi":"https://doi.org/10.24963/ijcai.2020/136","mag":"3182242813"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2020/136","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2020/136","pdf_url":"https://www.ijcai.org/proceedings/2020/0136.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.ijcai.org/proceedings/2020/0136.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5060456658","display_name":"Yuhui Xu","orcid":"https://orcid.org/0000-0002-7109-7140"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Yuhui Xu","raw_affiliation_strings":["Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076457803","display_name":"Yuxi Li","orcid":"https://orcid.org/0000-0002-6468-5454"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuxi Li","raw_affiliation_strings":["Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100328913","display_name":"Shuai Zhang","orcid":"https://orcid.org/0000-0002-9649-4740"},"institutions":[{"id":"https://openalex.org/I19268510","display_name":"Qualcomm (United Kingdom)","ror":"https://ror.org/04d3djg48","country_code":"GB","type":"company","lineage":["https://openalex.org/I19268510","https://openalex.org/I4210087596"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Shuai Zhang","raw_affiliation_strings":["Qualcomm AI Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Qualcomm AI Research","institution_ids":["https://openalex.org/I19268510"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101491962","display_name":"Wei Wen","orcid":"https://orcid.org/0009-0004-4662-7703"},"institutions":[{"id":"https://openalex.org/I170897317","display_name":"Duke University","ror":"https://ror.org/00py81415","country_code":"US","type":"education","lineage":["https://openalex.org/I170897317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wei Wen","raw_affiliation_strings":["Duke University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Duke University","institution_ids":["https://openalex.org/I170897317"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100606649","display_name":"Botao Wang","orcid":"https://orcid.org/0000-0003-2462-3088"},"institutions":[{"id":"https://openalex.org/I19268510","display_name":"Qualcomm (United Kingdom)","ror":"https://ror.org/04d3djg48","country_code":"GB","type":"company","lineage":["https://openalex.org/I19268510","https://openalex.org/I4210087596"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Botao Wang","raw_affiliation_strings":["Qualcomm AI Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Qualcomm AI Research","institution_ids":["https://openalex.org/I19268510"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101940576","display_name":"Yingyong Qi","orcid":"https://orcid.org/0000-0003-1430-258X"},"institutions":[{"id":"https://openalex.org/I19268510","display_name":"Qualcomm (United Kingdom)","ror":"https://ror.org/04d3djg48","country_code":"GB","type":"company","lineage":["https://openalex.org/I19268510","https://openalex.org/I4210087596"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Yingyong Qi","raw_affiliation_strings":["Qualcomm AI Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Qualcomm AI Research","institution_ids":["https://openalex.org/I19268510"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058073627","display_name":"Yiran Chen","orcid":"https://orcid.org/0000-0002-1486-8412"},"institutions":[{"id":"https://openalex.org/I170897317","display_name":"Duke University","ror":"https://ror.org/00py81415","country_code":"US","type":"education","lineage":["https://openalex.org/I170897317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yiran Chen","raw_affiliation_strings":["Duke University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Duke University","institution_ids":["https://openalex.org/I170897317"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059372830","display_name":"Weiyao Lin","orcid":"https://orcid.org/0000-0001-8307-7107"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Weiyao Lin","raw_affiliation_strings":["Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5002494284","display_name":"Hongkai Xiong","orcid":"https://orcid.org/0000-0003-4552-0029"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hongkai Xiong","raw_affiliation_strings":["Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":9,"corresponding_author_ids":["https://openalex.org/A5060456658"],"corresponding_institution_ids":["https://openalex.org/I183067930"],"apc_list":null,"apc_paid":null,"fwci":8.453,"has_fulltext":false,"cited_by_count":65,"citation_normalized_percentile":{"value":0.98961706,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"977","last_page":"983"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10500","display_name":"Sparse and Compressive Sensing Techniques","score":0.9997000098228455,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10500","display_name":"Sparse and Compressive Sensing Techniques","score":0.9997000098228455,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9995999932289124,"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/T10191","display_name":"Robotics and Sensor-Based Localization","score":0.9980999827384949,"subfield":{"id":"https://openalex.org/subfields/2202","display_name":"Aerospace Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.7386983633041382},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6605556607246399},{"id":"https://openalex.org/keywords/rank","display_name":"Rank (graph theory)","score":0.6566578149795532},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.5537494421005249},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.524369478225708},{"id":"https://openalex.org/keywords/low-rank-approximation","display_name":"Low-rank approximation","score":0.510913074016571},{"id":"https://openalex.org/keywords/approximation-algorithm","display_name":"Approximation algorithm","score":0.4563385546207428},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.4407896399497986},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.4241058826446533},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.38954654335975647},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.38287240266799927},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3231441378593445},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.22375383973121643}],"concepts":[{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.7386983633041382},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6605556607246399},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.6566578149795532},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.5537494421005249},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.524369478225708},{"id":"https://openalex.org/C90199385","wikidata":"https://www.wikidata.org/wiki/Q6692777","display_name":"Low-rank approximation","level":3,"score":0.510913074016571},{"id":"https://openalex.org/C148764684","wikidata":"https://www.wikidata.org/wiki/Q621751","display_name":"Approximation algorithm","level":2,"score":0.4563385546207428},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.4407896399497986},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4241058826446533},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.38954654335975647},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38287240266799927},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3231441378593445},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.22375383973121643},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","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},{"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/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C25023664","wikidata":"https://www.wikidata.org/wiki/Q1575637","display_name":"Hankel matrix","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.24963/ijcai.2020/136","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2020/136","pdf_url":"https://www.ijcai.org/proceedings/2020/0136.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.24963/ijcai.2020/136","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2020/136","pdf_url":"https://www.ijcai.org/proceedings/2020/0136.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G3438946646","display_name":null,"funder_award_id":"61932022","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G3628871441","display_name":null,"funder_award_id":"61720106001","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G4969694544","display_name":null,"funder_award_id":"17XD1401900","funder_id":"https://openalex.org/F4320335796","funder_display_name":"Program of Shanghai Academic Research Leader"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320335796","display_name":"Program of Shanghai Academic Research Leader","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3182242813.pdf","grobid_xml":"https://content.openalex.org/works/W3182242813.grobid-xml"},"referenced_works_count":29,"referenced_works":["https://openalex.org/W83404773","https://openalex.org/W1824659232","https://openalex.org/W1996901117","https://openalex.org/W2029213856","https://openalex.org/W2064980127","https://openalex.org/W2104636679","https://openalex.org/W2108598243","https://openalex.org/W2172166488","https://openalex.org/W2194775991","https://openalex.org/W2260663238","https://openalex.org/W2300242332","https://openalex.org/W2605135468","https://openalex.org/W2623451521","https://openalex.org/W2767827820","https://openalex.org/W2789142637","https://openalex.org/W2883111419","https://openalex.org/W2883780447","https://openalex.org/W2885340141","https://openalex.org/W2950248853","https://openalex.org/W2962851801","https://openalex.org/W2962965870","https://openalex.org/W2963000224","https://openalex.org/W2963163009","https://openalex.org/W2963363373","https://openalex.org/W2964228333","https://openalex.org/W2964233199","https://openalex.org/W2987878254","https://openalex.org/W3034999720","https://openalex.org/W3118608800"],"related_works":["https://openalex.org/W2373300491","https://openalex.org/W2395294869","https://openalex.org/W2378744544","https://openalex.org/W2964108162","https://openalex.org/W2772729003","https://openalex.org/W2962894463","https://openalex.org/W2962895005","https://openalex.org/W2894882275","https://openalex.org/W4237339638","https://openalex.org/W3115518109"],"abstract_inverted_index":{"To":[0],"enable":[1],"DNNs":[2],"on":[3,61,170],"edge":[4],"devices":[5],"like":[6],"mobile":[7],"phones,":[8],"low-rank":[9,35,75,118,146],"approximation":[10,39,76,87,105],"has":[11,144],"been":[12],"widely":[13],"adopted":[14],"because":[15],"of":[16,112],"its":[17],"solid":[18],"theoretical":[19],"rationale":[20],"and":[21,57,88,106,148,172],"efficient":[22],"implementations.":[23],"Several":[24],"previous":[25,80,175],"works":[26],"attempted":[27],"to":[28,65,73,132],"directly":[29],"approximate":[30],"a":[31,46,51,58,145],"pre-trained":[32],"model":[33],"by":[34,126],"decomposition;":[36],"however,":[37],"small":[38],"errors":[40],"in":[41,137],"parameters":[42],"can":[43],"ripple":[44],"over":[45],"large":[47],"prediction":[48],"loss.":[49],"As":[50],"result,":[52],"performance":[53,153],"usually":[54],"drops":[55],"significantly":[56],"sophisticated":[59],"effort":[60],"fine-tuning":[62,158],"is":[63,70,130,149,167],"required":[64],"recover":[66],"accuracy.":[67],"Apparently,":[68],"it":[69],"not":[71],"optimal":[72],"separate":[74],"from":[77],"training.":[78,107,121],"Unlike":[79],"works,":[81],"this":[82],"paper":[83],"integrates":[84],"low":[85,103,135,161,179],"rank":[86,104,136,162,180],"regularization":[89,124],"into":[90],"the":[91,110,113,157],"training":[92],"process.":[93],"We":[94],"propose":[95],"Trained":[96],"Rank":[97],"Pruning":[98],"(TRP),":[99],"which":[100],"alternates":[101],"between":[102],"TRP":[108,140],"maintains":[109],"capacity":[111],"original":[114],"network":[115,142],"while":[116],"imposing":[117],"constraints":[119],"during":[120],"A":[122],"nuclear":[123],"optimized":[125],"stochastic":[127],"sub-gradient":[128],"descent":[129],"utilized":[131],"further":[133],"promote":[134],"TRP.":[138],"The":[139,164],"trained":[141],"inherently":[143],"structure,":[147],"approximated":[150],"with":[151],"negligible":[152],"loss,":[154],"thus":[155],"eliminating":[156],"process":[159],"after":[160],"decomposition.":[163],"proposed":[165],"method":[166],"comprehensively":[168],"evaluated":[169],"CIFAR-10":[171],"ImageNet,":[173],"outperforming":[174],"compression":[176],"methods":[177],"using":[178],"approximation.":[181]},"counts_by_year":[{"year":2025,"cited_by_count":6},{"year":2024,"cited_by_count":10},{"year":2023,"cited_by_count":13},{"year":2022,"cited_by_count":10},{"year":2021,"cited_by_count":18},{"year":2020,"cited_by_count":6},{"year":2019,"cited_by_count":2}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}