{"id":"https://openalex.org/W4295832462","doi":"https://doi.org/10.1109/tcad.2022.3206394","title":"Accelerating On-Device DNN Training Workloads via Runtime Convergence Monitor","display_name":"Accelerating On-Device DNN Training Workloads via Runtime Convergence Monitor","publication_year":2022,"publication_date":"2022-09-13","ids":{"openalex":"https://openalex.org/W4295832462","doi":"https://doi.org/10.1109/tcad.2022.3206394"},"language":"en","primary_location":{"id":"doi:10.1109/tcad.2022.3206394","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2022.3206394","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/A5034558336","display_name":"Seungkyu Choi","orcid":"https://orcid.org/0000-0002-3125-9707"},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Seungkyu Choi","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Michigan, Ann Arbor, MI, USA"],"raw_orcid":"https://orcid.org/0000-0002-3125-9707","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Michigan, Ann Arbor, MI, USA","institution_ids":["https://openalex.org/I27837315"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072316644","display_name":"Jaekang Shin","orcid":"https://orcid.org/0000-0002-5943-1599"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jaekang Shin","raw_affiliation_strings":["School of Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, South Korea"],"raw_orcid":"https://orcid.org/0000-0002-5943-1599","affiliations":[{"raw_affiliation_string":"School of Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5052390471","display_name":"Lee\u2010Sup Kim","orcid":"https://orcid.org/0000-0001-9585-4591"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Lee-Sup Kim","raw_affiliation_strings":["School of Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, South Korea"],"raw_orcid":"https://orcid.org/0000-0001-9585-4591","affiliations":[{"raw_affiliation_string":"School of Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, South Korea","institution_ids":["https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.2437,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.46845115,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":96},"biblio":{"volume":"42","issue":"5","first_page":"1574","last_page":"1587"},"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9993000030517578,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9972000122070312,"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/computer-science","display_name":"Computer science","score":0.8503304719924927},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.8027479648590088},{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.5924073457717896},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.5183364152908325},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.5144693851470947},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5053661465644836},{"id":"https://openalex.org/keywords/adaptation","display_name":"Adaptation (eye)","score":0.5023612976074219},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4893204867839813},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.48473379015922546},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.4742909073829651},{"id":"https://openalex.org/keywords/convergence","display_name":"Convergence (economics)","score":0.4716975688934326},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.468699187040329},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4549711048603058},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4347483515739441},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.42326176166534424},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3100510239601135},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.23294398188591003},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.16109958291053772}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8503304719924927},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.8027479648590088},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.5924073457717896},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.5183364152908325},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.5144693851470947},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5053661465644836},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.5023612976074219},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4893204867839813},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.48473379015922546},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.4742909073829651},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.4716975688934326},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.468699187040329},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4549711048603058},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4347483515739441},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.42326176166534424},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3100510239601135},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.23294398188591003},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.16109958291053772},{"id":"https://openalex.org/C120665830","wikidata":"https://www.wikidata.org/wiki/Q14620","display_name":"Optics","level":1,"score":0.0},{"id":"https://openalex.org/C50522688","wikidata":"https://www.wikidata.org/wiki/Q189833","display_name":"Economic growth","level":1,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"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/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcad.2022.3206394","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2022.3206394","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":0.4000000059604645,"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure"}],"awards":[{"id":"https://openalex.org/G3807300858","display_name":null,"funder_award_id":"IITP-2020-0-01847","funder_id":"https://openalex.org/F4320328359","funder_display_name":"Ministry of Science and ICT, South Korea"},{"id":"https://openalex.org/G5939822931","display_name":null,"funder_award_id":"2020R1A2B5B02002690","funder_id":"https://openalex.org/F4320322030","funder_display_name":"Ministry of Science, ICT and Future Planning"}],"funders":[{"id":"https://openalex.org/F4320315121","display_name":"Samsung Advanced Institute of Technology","ror":null},{"id":"https://openalex.org/F4320322030","display_name":"Ministry of Science, ICT and Future Planning","ror":"https://ror.org/032e49973"},{"id":"https://openalex.org/F4320322120","display_name":"National Research Foundation of Korea","ror":"https://ror.org/013aysd81"},{"id":"https://openalex.org/F4320328359","display_name":"Ministry of Science and ICT, South Korea","ror":"https://ror.org/01wpjm123"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":38,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W2117539524","https://openalex.org/W2138011018","https://openalex.org/W2163605009","https://openalex.org/W2165698076","https://openalex.org/W2194775991","https://openalex.org/W2469490737","https://openalex.org/W2531409750","https://openalex.org/W2533598788","https://openalex.org/W2794478957","https://openalex.org/W2884360532","https://openalex.org/W2884367402","https://openalex.org/W2921918777","https://openalex.org/W2945992628","https://openalex.org/W2949866178","https://openalex.org/W2955425717","https://openalex.org/W2963163009","https://openalex.org/W2964332173","https://openalex.org/W2990765879","https://openalex.org/W2997972888","https://openalex.org/W3040850704","https://openalex.org/W3090288569","https://openalex.org/W3091941622","https://openalex.org/W3092345941","https://openalex.org/W3102587717","https://openalex.org/W3160017297","https://openalex.org/W3172967059","https://openalex.org/W3187235458","https://openalex.org/W3212514695","https://openalex.org/W4200223557","https://openalex.org/W4230981054","https://openalex.org/W6600609147","https://openalex.org/W6631190155","https://openalex.org/W6636751772","https://openalex.org/W6684191040","https://openalex.org/W6762718338","https://openalex.org/W6770166933","https://openalex.org/W6772018893"],"related_works":["https://openalex.org/W2319467001","https://openalex.org/W1784146144","https://openalex.org/W2086205578","https://openalex.org/W3196579076","https://openalex.org/W2898755250","https://openalex.org/W4296563896","https://openalex.org/W4295854938","https://openalex.org/W4286233754","https://openalex.org/W4323922040","https://openalex.org/W3216454443"],"abstract_inverted_index":{"With":[0],"the":[1,38,55,86,90,117,128,134,143,155,163,169,189,196],"growing":[2],"demand":[3],"for":[4,28,148,188,243],"processing":[5],"deep":[6],"learning":[7],"applications":[8],"on":[9,71,221],"edge":[10],"devices,":[11],"on-device":[12,119,244],"DNN":[13,91],"training":[14,39,73,92,120,135,149,170,187],"has":[15,95],"become":[16],"a":[17,22,107,176],"major":[18],"workload":[19],"to":[20,36,111,167,211],"execute":[21],"variety":[23],"of":[24,54,89,137,158,214],"vision":[25],"tasks":[26,139],"suited":[27],"users.":[29],"Therefore,":[30],"architectures":[31,233],"employed":[32],"with":[33,192],"algorithm":[34],"co-design":[35],"accelerate":[37],"process":[40,93],"have":[41],"been":[42,97],"steadily":[43],"studied.":[44],"However,":[45],"previous":[46,197],"solutions":[47],"are":[48,81],"mostly":[49],"supported":[50],"by":[51,162],"extended":[52],"versions":[53],"inference":[56],"studies,":[57],"such":[58],"as":[59],"sparsity,":[60],"data":[61],"flow,":[62],"quantization,":[63],"etc.":[64],"Moreover,":[65],"most":[66],"works":[67],"examine":[68],"their":[69],"schemes":[70],"from-the-scratch":[72],"that":[74,84,94],"cannot":[75],"tolerate":[76],"inaccurate":[77],"computing.":[78],"Accordingly,":[79],"there":[80],"still":[82],"factors":[83],"hinder":[85],"overall":[87],"speed":[88,171],"not":[96],"addressed":[98],"in":[99,116,145,185,230,240],"practical":[100,118],"workloads.":[101],"In":[102],"this":[103],"work,":[104],"we":[105,132,153,178],"propose":[106],"runtime":[108,204],"convergence":[109,144,184],"monitor":[110],"achieve":[112],"massive":[113],"computational":[114],"savings":[115],"workloads":[121],"(i.e.,":[122],"transfer-learning-based":[123],"task":[124,190],"adaptation).":[125],"By":[126],"monitoring":[127,201],"network":[129],"output":[130],"data,":[131],"determine":[133],"intensity":[136],"incoming":[138],"and":[140,206,234],"adaptively":[141],"detect":[142],"iteration":[146],"intervals":[147],"diverse":[150],"datasets.":[151],"Furthermore,":[152],"enable":[154],"computation":[156],"skip":[157],"converged":[159],"images":[160],"determined":[161],"monitored":[164],"prediction":[165],"probability":[166],"enhance":[168],"within":[172],"an":[173,180],"iteration.":[174],"As":[175],"result,":[177],"perform":[179],"accurate":[181],"but":[182],"fast":[183],"model":[186],"adaptation":[191],"minimal":[193],"overhead.":[194],"Unlike":[195],"approximation":[198],"methods,":[199],"our":[200],"system":[202],"enables":[203],"optimization":[205],"can":[207],"be":[208],"easily":[209],"applicable":[210],"any":[212,231],"type":[213],"accelerator":[215],"attaining":[216],"significant":[217],"speedup.":[218],"Evaluation":[219],"results":[220],"various":[222],"datasets":[223],"show":[224],"geomean":[225],"of$2.2\\times":[226],"$speedup":[227],"when":[228],"applied":[229,239],"systolic":[232],"further":[235],"enhancement":[236],"of$3.6\\times":[237],"$when":[238],"accelerators":[241],"dedicated":[242],"training.":[245]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1}],"updated_date":"2026-07-29T09:40:50.615796","created_date":"2025-10-10T00:00:00"}