{"id":"https://openalex.org/W4246550359","doi":"https://doi.org/10.1109/aspdac.2018.8297401","title":"CANNA: Neural network acceleration using configurable approximation on GPGPU","display_name":"CANNA: Neural network acceleration using configurable approximation on GPGPU","publication_year":2018,"publication_date":"2018-01-01","ids":{"openalex":"https://openalex.org/W4246550359","doi":"https://doi.org/10.1109/aspdac.2018.8297401"},"language":"en","primary_location":{"id":"doi:10.1109/aspdac.2018.8297401","is_oa":false,"landing_page_url":"https://doi.org/10.1109/aspdac.2018.8297401","pdf_url":null,"source":{"id":"https://openalex.org/S4363608266","display_name":"2018 23rd Asia and South Pacific Design Automation Conference (ASP-DAC)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 23rd Asia and South Pacific Design Automation Conference (ASP-DAC)","raw_type":"proceedings-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/A5033221192","display_name":"Mohsen Imani","orcid":"https://orcid.org/0000-0002-5761-0622"},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California, San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Mohsen Imani","raw_affiliation_strings":["CSE Department, UC San Diego, La Jolla, CA, USA"],"affiliations":[{"raw_affiliation_string":"CSE Department, UC San Diego, La Jolla, CA, USA","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088549329","display_name":"Max Masich","orcid":null},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California, San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Max Masich","raw_affiliation_strings":["CSE Department, UC San Diego, La Jolla, CA, USA"],"affiliations":[{"raw_affiliation_string":"CSE Department, UC San Diego, La Jolla, CA, USA","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003419054","display_name":"Daniel Peroni","orcid":null},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California, San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Daniel Peroni","raw_affiliation_strings":["CSE Department, UC San Diego, La Jolla, CA, USA"],"affiliations":[{"raw_affiliation_string":"CSE Department, UC San Diego, La Jolla, CA, USA","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061631231","display_name":"Pushen Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California, San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Pushen Wang","raw_affiliation_strings":["CSE Department, UC San Diego, La Jolla, CA, USA"],"affiliations":[{"raw_affiliation_string":"CSE Department, UC San Diego, La Jolla, CA, USA","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5112841571","display_name":"Tajana Rosing","orcid":null},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California, San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tajana Rosing","raw_affiliation_strings":["CSE Department, UC San Diego, La Jolla, CA, USA"],"affiliations":[{"raw_affiliation_string":"CSE Department, UC San Diego, La Jolla, CA, USA","institution_ids":["https://openalex.org/I36258959"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5033221192"],"corresponding_institution_ids":["https://openalex.org/I36258959"],"apc_list":null,"apc_paid":null,"fwci":1.4446,"has_fulltext":false,"cited_by_count":14,"citation_normalized_percentile":{"value":0.88149847,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":97},"biblio":{"volume":"22","issue":null,"first_page":"682","last_page":"689"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9995999932289124,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10320","display_name":"Neural Networks and Applications","score":0.9943000078201294,"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/speedup","display_name":"Speedup","score":0.8580273389816284},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7754570841789246},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6424123048782349},{"id":"https://openalex.org/keywords/canna","display_name":"Canna","score":0.5129894614219666},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4735794961452484},{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.44348978996276855},{"id":"https://openalex.org/keywords/graphics-processing-unit","display_name":"Graphics processing unit","score":0.43873918056488037},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.4339299201965332},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.43243151903152466},{"id":"https://openalex.org/keywords/approximation-error","display_name":"Approximation error","score":0.4246806502342224},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4120280146598816},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3760348856449127},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3341470956802368},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.24064892530441284},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.23133736848831177}],"concepts":[{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.8580273389816284},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7754570841789246},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6424123048782349},{"id":"https://openalex.org/C2777337493","wikidata":"https://www.wikidata.org/wiki/Q161182","display_name":"Canna","level":3,"score":0.5129894614219666},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4735794961452484},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.44348978996276855},{"id":"https://openalex.org/C2779851693","wikidata":"https://www.wikidata.org/wiki/Q183484","display_name":"Graphics processing unit","level":2,"score":0.43873918056488037},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.4339299201965332},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.43243151903152466},{"id":"https://openalex.org/C122383733","wikidata":"https://www.wikidata.org/wiki/Q865920","display_name":"Approximation error","level":2,"score":0.4246806502342224},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4120280146598816},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3760348856449127},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3341470956802368},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.24064892530441284},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.23133736848831177},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C529335014","wikidata":"https://www.wikidata.org/wiki/Q41534","display_name":"Starch","level":2,"score":0.0},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","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/C74650414","wikidata":"https://www.wikidata.org/wiki/Q11397","display_name":"Classical mechanics","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}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/aspdac.2018.8297401","is_oa":false,"landing_page_url":"https://doi.org/10.1109/aspdac.2018.8297401","pdf_url":null,"source":{"id":"https://openalex.org/S4363608266","display_name":"2018 23rd Asia and South Pacific Design Automation Conference (ASP-DAC)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 23rd Asia and South Pacific Design Automation Conference (ASP-DAC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.699999988079071,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":41,"referenced_works":["https://openalex.org/W104184427","https://openalex.org/W1991539813","https://openalex.org/W1998917233","https://openalex.org/W2048091247","https://openalex.org/W2095705004","https://openalex.org/W2103742924","https://openalex.org/W2105216798","https://openalex.org/W2121819644","https://openalex.org/W2148461049","https://openalex.org/W2161381512","https://openalex.org/W2163605009","https://openalex.org/W2286365479","https://openalex.org/W2346297475","https://openalex.org/W2346685430","https://openalex.org/W2352050502","https://openalex.org/W2396572963","https://openalex.org/W2499900168","https://openalex.org/W2520083297","https://openalex.org/W2529175623","https://openalex.org/W2533121491","https://openalex.org/W2537759401","https://openalex.org/W2586950459","https://openalex.org/W2587527900","https://openalex.org/W2611289746","https://openalex.org/W2613567208","https://openalex.org/W2625133503","https://openalex.org/W2625483679","https://openalex.org/W2725615981","https://openalex.org/W2753742176","https://openalex.org/W2950656546","https://openalex.org/W2964200805","https://openalex.org/W3141620748","https://openalex.org/W4230979891","https://openalex.org/W4242473664","https://openalex.org/W4297695444","https://openalex.org/W4361787200","https://openalex.org/W6604254268","https://openalex.org/W6674330103","https://openalex.org/W6677490153","https://openalex.org/W6681813608","https://openalex.org/W6684191040"],"related_works":["https://openalex.org/W2514371670","https://openalex.org/W2364486986","https://openalex.org/W2392304127","https://openalex.org/W2899948050","https://openalex.org/W4390047548","https://openalex.org/W4230669190","https://openalex.org/W4285281319","https://openalex.org/W4205705001","https://openalex.org/W1647216093","https://openalex.org/W2142903600"],"abstract_inverted_index":{"Neural":[0,23],"networks":[1,24],"have":[2],"been":[3],"successfully":[4],"used":[5],"in":[6,74,97,113,188],"many":[7],"applications.":[8],"Due":[9],"to":[10,17,85,143,166],"their":[11],"computational":[12],"complexity,":[13],"it":[14],"is":[15],"difficult":[16],"implement":[18],"them":[19,111],"on":[20,51,169],"embedded":[21],"devices.":[22],"are":[25],"inherently":[26],"approximate":[27],"and":[28,109,120,148,186],"thus":[29],"can":[30],"be":[31],"simplified.":[32],"In":[33],"this":[34],"paper,":[35],"CANNA":[36,140],"proposes":[37],"a":[38,81,92],"gradual":[39],"training":[40,153],"approximation":[41,49,68,107],"which":[42,103],"adaptively":[43],"sets":[44],"the":[45,52,72,105,118,167,173,180],"level":[46],"of":[47,58,77,122],"hardware":[48,61,88],"depending":[50],"neural":[53,78,156],"network's":[54],"internal":[55],"error,":[56],"instead":[57,112],"apply":[59],"uniform":[60],"approximation.":[62,86],"To":[63],"accelerate":[64],"inference,":[65],"CANNA's":[66],"layer-based":[67,177],"approach":[69,178],"selectively":[70],"relaxes":[71],"computation":[73],"each":[75],"layer":[76],"network,":[79],"as":[80,164],"function":[82],"its":[83],"sensitivity":[84],"For":[87],"support,":[89],"we":[90],"use":[91],"configurable":[93,127],"floating":[94],"point":[95],"unit":[96],"Hardware":[98],"that":[99,139],"dynamically":[100],"identifies":[101],"inputs":[102],"produce":[104],"largest":[106],"error":[108],"process":[110],"precise":[114],"mode.":[115],"We":[116],"evaluate":[117],"accuracy":[119],"efficiency":[121,182],"our":[123,176],"design":[124],"by":[125,183],"integrating":[126],"FPUs":[128],"into":[129],"AMD's":[130],"Southern":[131],"Island":[132],"GPU":[133],"architecture.":[134],"Our":[135],"experimental":[136],"evaluation":[137],"shows":[138],"achieves":[141],"up":[142],"4.84\u00d7":[144],"(7.13\u00d7)":[145],"energy":[146,181],"savings":[147],"3.22\u00d7":[149],"(4.64\u00d7)":[150],"speedup":[151,191],"when":[152],"four":[154],"different":[155],"network":[157],"applications":[158],"with":[159],"0%":[160,194],"(2%)":[161,195],"quality":[162,196],"loss":[163],"compared":[165],"implementation":[168],"baseline":[170],"GPU.":[171],"During":[172],"inference":[174],"phase,":[175],"improves":[179],"4.42\u00d7":[184],"(6.06\u00d7)":[185],"results":[187],"2.96\u00d7":[189],"(3.98\u00d7)":[190],"while":[192],"ensuring":[193],"loss.":[197]},"counts_by_year":[{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":4},{"year":2020,"cited_by_count":4},{"year":2019,"cited_by_count":2},{"year":2018,"cited_by_count":2}],"updated_date":"2026-04-05T17:49:38.594831","created_date":"2025-10-10T00:00:00"}