{"id":"https://openalex.org/W3092345941","doi":"https://doi.org/10.1109/dac18072.2020.9218650","title":"Prediction Confidence based Low Complexity Gradient Computation for Accelerating DNN Training","display_name":"Prediction Confidence based Low Complexity Gradient Computation for Accelerating DNN Training","publication_year":2020,"publication_date":"2020-07-01","ids":{"openalex":"https://openalex.org/W3092345941","doi":"https://doi.org/10.1109/dac18072.2020.9218650","mag":"3092345941"},"language":"en","primary_location":{"id":"doi:10.1109/dac18072.2020.9218650","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac18072.2020.9218650","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 57th ACM/IEEE Design Automation Conference (DAC)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5091058036","display_name":"Dongyeob Shin","orcid":null},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Dongyeob Shin","raw_affiliation_strings":["School of Electrical Engineering, Korea University, Seoul, Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Electrical Engineering, Korea University, Seoul, Korea","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101771812","display_name":"Geonho Kim","orcid":"https://orcid.org/0000-0003-0587-1975"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Geonho Kim","raw_affiliation_strings":["School of Electrical Engineering, Korea University, Seoul, Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Electrical Engineering, Korea University, Seoul, Korea","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035754102","display_name":"Joongho Jo","orcid":null},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Joongho Jo","raw_affiliation_strings":["School of Electrical Engineering, Korea University, Seoul, Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Electrical Engineering, Korea University, Seoul, Korea","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101839916","display_name":"Jongsun Park","orcid":"https://orcid.org/0000-0003-3251-0024"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jongsun Park","raw_affiliation_strings":["School of Electrical Engineering, Korea University, Seoul, Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Electrical Engineering, Korea University, Seoul, Korea","institution_ids":["https://openalex.org/I197347611"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I197347611"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":14,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9990000128746033,"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":0.9984999895095825,"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/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.7594684958457947},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.7504302263259888},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6754046678543091},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.6280797719955444},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5705839395523071},{"id":"https://openalex.org/keywords/confidence-interval","display_name":"Confidence interval","score":0.4694373607635498},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.45821231603622437},{"id":"https://openalex.org/keywords/backpropagation","display_name":"Backpropagation","score":0.4230685532093048},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3415825068950653},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.20185208320617676},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1697780191898346}],"concepts":[{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.7594684958457947},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.7504302263259888},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6754046678543091},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.6280797719955444},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5705839395523071},{"id":"https://openalex.org/C44249647","wikidata":"https://www.wikidata.org/wiki/Q208498","display_name":"Confidence interval","level":2,"score":0.4694373607635498},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.45821231603622437},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.4230685532093048},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3415825068950653},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.20185208320617676},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1697780191898346}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/dac18072.2020.9218650","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac18072.2020.9218650","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 57th ACM/IEEE Design Automation Conference (DAC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.9100000262260437,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":21,"referenced_works":["https://openalex.org/W114517082","https://openalex.org/W2194775991","https://openalex.org/W2407022425","https://openalex.org/W2523246573","https://openalex.org/W2617766261","https://openalex.org/W2884360532","https://openalex.org/W2888885397","https://openalex.org/W2890616119","https://openalex.org/W2891800530","https://openalex.org/W2905102070","https://openalex.org/W2913839334","https://openalex.org/W2944850220","https://openalex.org/W2945992628","https://openalex.org/W2970472742","https://openalex.org/W2971297906","https://openalex.org/W4289548082","https://openalex.org/W6727249380","https://openalex.org/W6738460352","https://openalex.org/W6754171319","https://openalex.org/W6754799804","https://openalex.org/W6759163087"],"related_works":["https://openalex.org/W2115605526","https://openalex.org/W3093883775","https://openalex.org/W1539246760","https://openalex.org/W2786746258","https://openalex.org/W4225893763","https://openalex.org/W2788727425","https://openalex.org/W2405196115","https://openalex.org/W2104893957","https://openalex.org/W2971074373","https://openalex.org/W4402471162"],"abstract_inverted_index":{"In":[0,46],"deep":[1],"neural":[2],"network":[3,6,71],"(DNN)":[4],"training,":[5],"weights":[7],"are":[8,16],"iteratively":[9],"updated":[10],"with":[11,29,74,147],"the":[12,55,67,77,80,83,94,111,121],"weight":[13,32,56,86],"gradients":[14],"that":[15,66,128,152],"obtained":[17],"from":[18],"stochastic":[19],"gradient":[20,27,33,57,87,95,135],"descent":[21],"(SGD).":[22],"Since":[23],"SGD":[24],"inherently":[25],"allows":[26],"calculations":[28],"noise,":[30],"approximating":[31],"computations":[34,96,136],"have":[35],"a":[36,106],"large":[37],"potential":[38],"of":[39,54,62,70,85,134,159],"training":[40,63,75,163,177],"energy/time":[41],"savings":[42],"without":[43,97,142],"degrading":[44],"accuracy.":[45],"this":[47],"paper,":[48],"we":[49],"propose":[50],"an":[51],"input-dependent":[52],"approximation":[53],"for":[58,101,137,169],"improving":[59],"energy":[60,164],"efficiency":[61],"process.":[64],"Considering":[65],"output":[68],"predictions":[69],"(confidence)":[72],"changes":[73],"inputs,":[76],"relation":[78],"between":[79],"confidence":[81,103,122],"and":[82,157,165],"magnitude":[84],"can":[88,115,131],"be":[89,116],"efficiently":[90],"exploited":[91],"to":[92,175],"skip":[93,113,132],"accuracy":[98,143],"drop,":[99],"especially":[100],"high":[102],"inputs.":[104],"With":[105],"given":[107],"squared":[108],"error":[109],"constraint,":[110],"computation":[112],"rates":[114],"also":[117],"controlled":[118],"by":[119],"changing":[120],"threshold.":[123],"The":[124],"simulation":[125],"results":[126],"show":[127],"our":[129,153],"approach":[130],"72.6%":[133],"CIFAR-100":[138,170],"dataset":[139,171],"using":[140,172],"ResNet-18":[141,173],"degradation.":[144],"Hardware":[145],"implementation":[146],"65nm":[148],"CMOS":[149],"process":[150],"shows":[151],"design":[154],"achieves":[155],"88.84%":[156],"98.16%":[158],"maximum":[160],"per":[161],"epoch":[162],"time":[166],"savings,":[167],"respectively,":[168],"compared":[174],"state-of-the-art":[176],"accelerator.":[178]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":4},{"year":2022,"cited_by_count":5},{"year":2021,"cited_by_count":3},{"year":2020,"cited_by_count":1}],"updated_date":"2026-07-15T18:14:33.161393","created_date":"2025-10-10T00:00:00"}