{"id":"https://openalex.org/W3005828390","doi":"https://doi.org/10.1109/dac18072.2020.9218596","title":"Best of Both Worlds: AutoML Codesign of a CNN and its Hardware Accelerator","display_name":"Best of Both Worlds: AutoML Codesign of a CNN and its Hardware Accelerator","publication_year":2020,"publication_date":"2020-07-01","ids":{"openalex":"https://openalex.org/W3005828390","doi":"https://doi.org/10.1109/dac18072.2020.9218596","mag":"3005828390"},"language":"en","primary_location":{"id":"doi:10.1109/dac18072.2020.9218596","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac18072.2020.9218596","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":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2002.05022","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5010814272","display_name":"Mohamed S. Abdelfattah","orcid":"https://orcid.org/0000-0002-4568-8932"},"institutions":[{"id":"https://openalex.org/I4210101778","display_name":"Samsung (United States)","ror":"https://ror.org/01bfbvm65","country_code":"US","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210101778"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mohamed S. Abdelfattah","raw_affiliation_strings":["Samsung AI Center, Cambridge","[Samsung]"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung AI Center, Cambridge","institution_ids":["https://openalex.org/I4210101778"]},{"raw_affiliation_string":"[Samsung]","institution_ids":["https://openalex.org/I4210101778"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080541362","display_name":"\u0141ukasz Dudziak","orcid":"https://orcid.org/0000-0003-4929-265X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lukasz Dudziak","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110651248","display_name":"Thomas Chau","orcid":null},"institutions":[{"id":"https://openalex.org/I4210101778","display_name":"Samsung (United States)","ror":"https://ror.org/01bfbvm65","country_code":"US","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210101778"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Thomas Chau","raw_affiliation_strings":["Samsung AI Center, Cambridge"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung AI Center, Cambridge","institution_ids":["https://openalex.org/I4210101778"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026267258","display_name":"Royson Lee","orcid":"https://orcid.org/0000-0002-6716-7994"},"institutions":[{"id":"https://openalex.org/I4210101778","display_name":"Samsung (United States)","ror":"https://ror.org/01bfbvm65","country_code":"US","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210101778"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Royson Lee","raw_affiliation_strings":["Samsung AI Center, Cambridge"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung AI Center, Cambridge","institution_ids":["https://openalex.org/I4210101778"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072204319","display_name":"Hyeji Kim","orcid":"https://orcid.org/0000-0002-2256-5729"},"institutions":[{"id":"https://openalex.org/I4210101778","display_name":"Samsung (United States)","ror":"https://ror.org/01bfbvm65","country_code":"US","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210101778"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hyeji Kim","raw_affiliation_strings":["Samsung AI Center, Cambridge"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung AI Center, Cambridge","institution_ids":["https://openalex.org/I4210101778"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5045638679","display_name":"Nicholas D. Lane","orcid":"https://orcid.org/0000-0002-2728-8273"},"institutions":[{"id":"https://openalex.org/I40120149","display_name":"University of Oxford","ror":"https://ror.org/052gg0110","country_code":"GB","type":"education","lineage":["https://openalex.org/I40120149"]},{"id":"https://openalex.org/I4210101778","display_name":"Samsung (United States)","ror":"https://ror.org/01bfbvm65","country_code":"US","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210101778"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Nicholas D. Lane","raw_affiliation_strings":["Samsung AI Center, Cambridge","University of Oxford"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung AI Center, Cambridge","institution_ids":["https://openalex.org/I4210101778"]},{"raw_affiliation_string":"University of Oxford","institution_ids":["https://openalex.org/I40120149"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":15,"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":0.9998000264167786,"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":0.9998000264167786,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9994999766349792,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.9976999759674072,"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":0.7662448883056641},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.736320436000824},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.5723362565040588},{"id":"https://openalex.org/keywords/pareto-principle","display_name":"Pareto principle","score":0.4993429183959961},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.48825764656066895},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.47346916794776917},{"id":"https://openalex.org/keywords/space","display_name":"Space (punctuation)","score":0.4363245964050293},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.41775041818618774},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.41638779640197754},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.41437965631484985},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.4137803912162781},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.4134214222431183},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.33985841274261475},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.335890531539917},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.32778364419937134},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.14720892906188965},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.10384738445281982},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.07881879806518555}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7662448883056641},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.736320436000824},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.5723362565040588},{"id":"https://openalex.org/C137635306","wikidata":"https://www.wikidata.org/wiki/Q182667","display_name":"Pareto principle","level":2,"score":0.4993429183959961},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.48825764656066895},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47346916794776917},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.4363245964050293},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.41775041818618774},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.41638779640197754},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.41437965631484985},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.4137803912162781},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.4134214222431183},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.33985841274261475},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.335890531539917},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.32778364419937134},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.14720892906188965},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.10384738445281982},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.07881879806518555},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1109/dac18072.2020.9218596","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac18072.2020.9218596","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"},{"id":"pmh:oai:arXiv.org:2002.05022","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2002.05022","pdf_url":"https://arxiv.org/pdf/2002.05022","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"mag:3005828390","is_oa":true,"landing_page_url":"http://export.arxiv.org/pdf/2002.05022","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.2002.05022","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2002.05022","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2002.05022","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2002.05022","pdf_url":"https://arxiv.org/pdf/2002.05022","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3005828390.pdf","grobid_xml":"https://content.openalex.org/works/W3005828390.grobid-xml"},"referenced_works_count":22,"referenced_works":["https://openalex.org/W2082295213","https://openalex.org/W2097117768","https://openalex.org/W2194775991","https://openalex.org/W2553303224","https://openalex.org/W2606722458","https://openalex.org/W2883929540","https://openalex.org/W2912521296","https://openalex.org/W2913541967","https://openalex.org/W2916118939","https://openalex.org/W2945146780","https://openalex.org/W2953212265","https://openalex.org/W2962856739","https://openalex.org/W2963511748","https://openalex.org/W2963918968","https://openalex.org/W2972736078","https://openalex.org/W3016142271","https://openalex.org/W6674914833","https://openalex.org/W6729956949","https://openalex.org/W6758657797","https://openalex.org/W6759278257","https://openalex.org/W6759402996","https://openalex.org/W6765522929"],"related_works":["https://openalex.org/W3092216346","https://openalex.org/W2963918968","https://openalex.org/W2964259004","https://openalex.org/W3092334294","https://openalex.org/W2962856739","https://openalex.org/W2279098554","https://openalex.org/W3016142271","https://openalex.org/W2967733054","https://openalex.org/W2606722458","https://openalex.org/W2553303224","https://openalex.org/W2289252105","https://openalex.org/W2067523571","https://openalex.org/W3200067639","https://openalex.org/W2965658867","https://openalex.org/W2946948417","https://openalex.org/W2940862705","https://openalex.org/W2810075754","https://openalex.org/W2612445135","https://openalex.org/W2194775991","https://openalex.org/W2094756095"],"abstract_inverted_index":{"Neural":[0],"architecture":[1],"search":[2,70,108,130,140],"(NAS)":[3],"has":[4],"been":[5],"very":[6],"successful":[7],"at":[8],"outperforming":[9],"human-designed":[10],"convolutional":[11],"neural":[12],"networks":[13],"(CNN)":[14],"in":[15,172],"accuracy,":[16],"and":[17,63,67,79,100,122],"when":[18],"hardware":[19,43],"information":[20],"is":[21],"present,":[22],"latency":[23],"as":[24],"well.":[25],"However,":[26],"NAS-designed":[27],"CNNs":[28],"typically":[29],"have":[30],"a":[31,41],"complicated":[32],"topology,":[33],"therefore,":[34],"it":[35],"may":[36],"be":[37],"difficult":[38],"to":[39,117,135,144],"design":[40,156],"custom":[42],"(HW)":[44],"accelerator":[45,151],"for":[46,71],"such":[47],"CNNs.":[48],"We":[49,81],"automate":[50],"HW-CNN":[51],"codesign":[52,107],"using":[53],"NAS":[54],"by":[55,164,170],"including":[56],"parameters":[57],"from":[58,152],"both":[59],"the":[60,64,72,92,106,124],"CNN":[61],"model":[62],"HW":[65,150,155],"accelerator,":[66],"we":[68,88,114,158],"jointly":[69],"best":[73],"model-accelerator":[74,120],"pair":[75],"that":[76,128],"boosts":[77],"accuracy":[78,163],"efficiency.":[80],"call":[82],"this":[83,86],"Codesign-NAS.":[84,178],"In":[85],"paper":[87],"focus":[89],"on":[90,146,160],"defining":[91],"Codesign-NAS":[93],"multiobjective":[94],"optimization":[95],"problem,":[96],"demonstrating":[97],"its":[98,147],"effectiveness,":[99],"exploring":[101],"different":[102,138],"ways":[103],"of":[104,176],"navigating":[105],"space.":[109,131],"For":[110],"CIFAR-10":[111],"image":[112],"classification,":[113],"enumerate":[115],"close":[116],"4":[118],"billion":[119],"pairs,":[121],"find":[123],"Pareto":[125],"frontier":[126],"within":[127,153],"large":[129],"This":[132],"allows":[133],"us":[134],"evaluate":[136],"three":[137],"reinforcement-learning-based":[139],"strategies.":[141],"Finally,":[142],"compared":[143],"ResNet":[145],"most":[148],"optimal":[149],"our":[154],"space,":[157],"improve":[159],"CIFAR-100":[161],"classification":[162],"1.3%":[165],"while":[166],"simultaneously":[167],"increasing":[168],"performance/area":[169],"41%":[171],"just":[173],"~1000":[174],"GPU-hours":[175],"running":[177]},"counts_by_year":[{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":12},{"year":2020,"cited_by_count":2}],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2025-10-10T00:00:00"}