{"id":"https://openalex.org/W3090288569","doi":"https://doi.org/10.1109/tcad.2020.3012216","title":"Enabling On-Device CNN Training by Self-Supervised Instance Filtering and Error Map Pruning","display_name":"Enabling On-Device CNN Training by Self-Supervised Instance Filtering and Error Map Pruning","publication_year":2020,"publication_date":"2020-10-02","ids":{"openalex":"https://openalex.org/W3090288569","doi":"https://doi.org/10.1109/tcad.2020.3012216","mag":"3090288569"},"language":"en","primary_location":{"id":"doi:10.1109/tcad.2020.3012216","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2020.3012216","pdf_url":null,"source":{"id":"https://openalex.org/S100835903","display_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","issn_l":"0278-0070","issn":["0278-0070","1937-4151"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["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 Computer-Aided Design of Integrated Circuits and Systems","raw_type":"journal-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/A5019020335","display_name":"Yawen Wu","orcid":"https://orcid.org/0000-0001-6840-267X"},"institutions":[{"id":"https://openalex.org/I170201317","display_name":"University of Pittsburgh","ror":"https://ror.org/01an3r305","country_code":"US","type":"education","lineage":["https://openalex.org/I170201317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yawen Wu","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA"],"raw_orcid":"https://orcid.org/0000-0001-6840-267X","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I170201317"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002340105","display_name":"Zhepeng Wang","orcid":"https://orcid.org/0000-0002-1460-1128"},"institutions":[{"id":"https://openalex.org/I170201317","display_name":"University of Pittsburgh","ror":"https://ror.org/01an3r305","country_code":"US","type":"education","lineage":["https://openalex.org/I170201317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhepeng Wang","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA"],"raw_orcid":"https://orcid.org/0000-0002-1460-1128","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I170201317"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000141831","display_name":"Yiyu Shi","orcid":"https://orcid.org/0000-0002-6788-9823"},"institutions":[{"id":"https://openalex.org/I107639228","display_name":"University of Notre Dame","ror":"https://ror.org/00mkhxb43","country_code":"US","type":"education","lineage":["https://openalex.org/I107639228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yiyu Shi","raw_affiliation_strings":["Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN, USA","institution_ids":["https://openalex.org/I107639228"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5066534595","display_name":"Jingtong Hu","orcid":"https://orcid.org/0000-0003-4029-4034"},"institutions":[{"id":"https://openalex.org/I170201317","display_name":"University of Pittsburgh","ror":"https://ror.org/01an3r305","country_code":"US","type":"education","lineage":["https://openalex.org/I170201317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jingtong Hu","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA"],"raw_orcid":"https://orcid.org/0000-0003-4029-4034","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I170201317"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.3958,"has_fulltext":false,"cited_by_count":29,"citation_normalized_percentile":{"value":0.9085061,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":90,"max":99},"biblio":{"volume":"39","issue":"11","first_page":"3445","last_page":"3457"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":1.0,"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":1.0,"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":0.9966999888420105,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.995199978351593,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8172056674957275},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.7981849312782288},{"id":"https://openalex.org/keywords/mnist-database","display_name":"MNIST database","score":0.7563823461532593},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.628960371017456},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.6097787618637085},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.577433705329895},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.5410481691360474},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.5090479254722595},{"id":"https://openalex.org/keywords/boosting","display_name":"Boosting (machine learning)","score":0.4803406000137329},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4529092609882355},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.43809157609939575},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.41733318567276},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.2516590356826782},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.0869779884815216}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8172056674957275},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.7981849312782288},{"id":"https://openalex.org/C190502265","wikidata":"https://www.wikidata.org/wiki/Q17069496","display_name":"MNIST database","level":3,"score":0.7563823461532593},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.628960371017456},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.6097787618637085},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.577433705329895},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.5410481691360474},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.5090479254722595},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.4803406000137329},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4529092609882355},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.43809157609939575},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.41733318567276},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2516590356826782},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.0869779884815216},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","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/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"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/tcad.2020.3012216","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2020.3012216","pdf_url":null,"source":{"id":"https://openalex.org/S100835903","display_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","issn_l":"0278-0070","issn":["0278-0070","1937-4151"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["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 Computer-Aided Design of Integrated Circuits and Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.8500000238418579,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":45,"referenced_works":["https://openalex.org/W1686810756","https://openalex.org/W2108598243","https://openalex.org/W2119144962","https://openalex.org/W2139047213","https://openalex.org/W2194775991","https://openalex.org/W2267244539","https://openalex.org/W2321533354","https://openalex.org/W2331143823","https://openalex.org/W2546536770","https://openalex.org/W2739068567","https://openalex.org/W2767827820","https://openalex.org/W2769527235","https://openalex.org/W2769856846","https://openalex.org/W2785722081","https://openalex.org/W2794478957","https://openalex.org/W2803113791","https://openalex.org/W2806285385","https://openalex.org/W2884711234","https://openalex.org/W2885765965","https://openalex.org/W2895298874","https://openalex.org/W2900120080","https://openalex.org/W2911884654","https://openalex.org/W2912283611","https://openalex.org/W2949899225","https://openalex.org/W2950745363","https://openalex.org/W2963163009","https://openalex.org/W2963516811","https://openalex.org/W2963709416","https://openalex.org/W2964152344","https://openalex.org/W2967733054","https://openalex.org/W2979542579","https://openalex.org/W2984989518","https://openalex.org/W2988851551","https://openalex.org/W2990765879","https://openalex.org/W3013688681","https://openalex.org/W3016142271","https://openalex.org/W3018620046","https://openalex.org/W3092581393","https://openalex.org/W3100470991","https://openalex.org/W4230714977","https://openalex.org/W4292084264","https://openalex.org/W4318619660","https://openalex.org/W6741753795","https://openalex.org/W6745670897","https://openalex.org/W6751528251"],"related_works":["https://openalex.org/W2950475743","https://openalex.org/W4386603768","https://openalex.org/W2886711096","https://openalex.org/W3036048022","https://openalex.org/W4309224979","https://openalex.org/W992687842","https://openalex.org/W4300560302","https://openalex.org/W3088108839","https://openalex.org/W3026879719","https://openalex.org/W3166492421"],"abstract_inverted_index":{"This":[0],"work":[1],"aims":[2],"to":[3,35,133],"enable":[4],"on-device":[5],"training":[6,18,60,79,94,149,176,201,229,239],"of":[7,198],"convolutional":[8],"neural":[9],"networks":[10],"(CNNs)":[11],"by":[12,61,100,212],"reducing":[13],"the":[14,30,39,64,74,90,97,119,130,135,151,158],"computation":[15,76,98,159,183,191,208],"cost":[16,77,99,162],"at":[17],"time.":[19],"CNN":[20],"models":[21,32],"are":[22,33],"usually":[23],"trained":[24,31,41],"on":[25,109,118,178,203,220,231,241],"high-performance":[26],"computers":[27],"and":[28,50,95,112,137,160,189,234],"only":[29],"deployed":[34],"edge":[36,221],"devices.":[37,83],"But":[38],"statically":[40],"model":[42],"cannot":[43],"adapt":[44],"dynamically":[45],"in":[46,53,93],"a":[47,194],"real":[48],"environment":[49],"may":[51],"result":[52],"low":[54],"accuracy":[55,171,188,196,211,245],"for":[56,81],"new":[57],"inputs.":[58],"On-device":[59],"learning":[62],"from":[63,129],"real-world":[65],"data":[66,110],"after":[67],"deployment":[68],"can":[69],"greatly":[70],"improve":[71],"accuracy.":[72],"However,":[73],"high":[75],"makes":[78],"prohibitive":[80],"resource-constrained":[82],"To":[84],"tackle":[85],"this":[86],"problem,":[87],"we":[88,180,205],"explore":[89],"computational":[91],"redundancies":[92],"reduce":[96],"two":[101],"complementary":[102],"approaches:":[103],"1)":[104],"self-supervised":[105],"early":[106,123],"instance":[107,124],"filtering":[108],"level":[111],"2)":[113],"error":[114],"map":[115],"pruning":[116],"(EMP)":[117],"algorithm":[120],"level.":[121],"The":[122,141],"filter":[125],"selects":[126],"important":[127],"instances":[128],"input":[131],"stream":[132],"train":[134],"network":[136],"drops":[138],"trivial":[139],"ones.":[140],"EMP":[142],"further":[143],"prunes":[144],"out":[145],"insignificant":[146],"computations":[147],"when":[148,175,228,238],"with":[150,169,193],"selected":[152],"instances.":[153],"Extensive":[154],"experiments":[155],"show":[156],"that":[157],"energy":[161,216,226,236],"is":[163,218],"substantially":[164],"reduced":[165],"without":[166,243],"any":[167,244],"or":[168],"marginal":[170,195],"loss.":[172,246],"For":[173],"example,":[174],"ResNet-110":[177,230],"CIFAR-10,":[179],"achieve":[181,224],"67.8%":[182],"saving":[184,192,217,227,237],"while":[185,209],"preserving":[186],"full":[187],"75.1%":[190],"loss":[197],"1.3%.":[199],"When":[200],"LeNet":[202,240],"MNIST,":[204],"save":[206],"79%":[207],"boosting":[210],"0.2%.":[213],"Besides,":[214],"practical":[215],"measured":[219],"platforms.":[222],"We":[223],"67.6%":[225],"mobile":[232],"GPU":[233],"74.1%":[235],"MCU":[242]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":9},{"year":2022,"cited_by_count":8},{"year":2021,"cited_by_count":6},{"year":2020,"cited_by_count":2}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}