{"id":"https://openalex.org/W2896983500","doi":"https://doi.org/10.1109/tcad.2017.2785257","title":"Caffeine: Toward Uniformed Representation and Acceleration for Deep Convolutional Neural Networks","display_name":"Caffeine: Toward Uniformed Representation and Acceleration for Deep Convolutional Neural Networks","publication_year":2018,"publication_date":"2018-10-18","ids":{"openalex":"https://openalex.org/W2896983500","doi":"https://doi.org/10.1109/tcad.2017.2785257","mag":"2896983500"},"language":"en","primary_location":{"id":"doi:10.1109/tcad.2017.2785257","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2017.2785257","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/A5100374175","display_name":"Chen Zhang","orcid":"https://orcid.org/0000-0003-2762-2726"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Chen Zhang","raw_affiliation_strings":["Center for Energy-Efficient Computing and Applications, Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Center for Energy-Efficient Computing and Applications, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101850376","display_name":"Guangyu Sun","orcid":"https://orcid.org/0000-0001-6436-0820"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guangyu Sun","raw_affiliation_strings":["Center for Energy-Efficient Computing and Applications, Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Center for Energy-Efficient Computing and Applications, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065889904","display_name":"Zhenman Fang","orcid":"https://orcid.org/0000-0003-0603-9697"},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhenman Fang","raw_affiliation_strings":["Center for Domain-Specific Computing, University of California at Los Angeles, Los Angeles, CA, USA"],"affiliations":[{"raw_affiliation_string":"Center for Domain-Specific Computing, University of California at Los Angeles, Los Angeles, CA, USA","institution_ids":["https://openalex.org/I161318765"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063866156","display_name":"Peipei Zhou","orcid":"https://orcid.org/0000-0002-0493-1844"},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Peipei Zhou","raw_affiliation_strings":["Center for Domain-Specific Computing, University of California at Los Angeles, Los Angeles, CA, USA"],"affiliations":[{"raw_affiliation_string":"Center for Domain-Specific Computing, University of California at Los Angeles, Los Angeles, CA, USA","institution_ids":["https://openalex.org/I161318765"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069821335","display_name":"Peichen Pan","orcid":"https://orcid.org/0000-0003-2947-8991"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peichen Pan","raw_affiliation_strings":["Falcon Computing Solutions Inc., Los Angeles, CA, USA"],"affiliations":[{"raw_affiliation_string":"Falcon Computing Solutions Inc., Los Angeles, CA, USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5016776689","display_name":"Jason Cong","orcid":"https://orcid.org/0000-0003-2887-6963"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jason Cong","raw_affiliation_strings":["Center for Energy-Efficient Computing and Applications, Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Center for Energy-Efficient Computing and Applications, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5100374175"],"corresponding_institution_ids":["https://openalex.org/I20231570"],"apc_list":null,"apc_paid":null,"fwci":14.341,"has_fulltext":false,"cited_by_count":265,"citation_normalized_percentile":{"value":0.99111016,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":100},"biblio":{"volume":"38","issue":"11","first_page":"2072","last_page":"2085"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998999834060669,"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":0.9998999834060669,"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9969000220298767,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9952999949455261,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8196008205413818},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.7695844173431396},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.7273775339126587},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.7221900224685669},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.6331626176834106},{"id":"https://openalex.org/keywords/xeon","display_name":"Xeon","score":0.5356471538543701},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5243085622787476},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.4505482316017151},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.435375839471817},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.42643776535987854},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4104463458061218},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3626731038093567}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8196008205413818},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.7695844173431396},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.7273775339126587},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.7221900224685669},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6331626176834106},{"id":"https://openalex.org/C145108525","wikidata":"https://www.wikidata.org/wiki/Q656154","display_name":"Xeon","level":2,"score":0.5356471538543701},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5243085622787476},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.4505482316017151},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.435375839471817},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.42643776535987854},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4104463458061218},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3626731038093567}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcad.2017.2785257","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2017.2785257","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":[{"display_name":"Affordable and clean energy","score":0.8700000047683716,"id":"https://metadata.un.org/sdg/7"}],"awards":[{"id":"https://openalex.org/G1659857404","display_name":null,"funder_award_id":"61572045","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320309469","display_name":"Mentor Graphics","ror":"https://ror.org/00r59k631"},{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320322725","display_name":"China Scholarship Council","ror":"https://ror.org/04atp4p48"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":45,"referenced_works":["https://openalex.org/W1677182931","https://openalex.org/W1686810756","https://openalex.org/W1849277567","https://openalex.org/W1970088388","https://openalex.org/W1982052956","https://openalex.org/W1983364832","https://openalex.org/W1990315422","https://openalex.org/W2002555321","https://openalex.org/W2009832130","https://openalex.org/W2012897754","https://openalex.org/W2015048865","https://openalex.org/W2053968820","https://openalex.org/W2094756095","https://openalex.org/W2096645269","https://openalex.org/W2097117768","https://openalex.org/W2102605133","https://openalex.org/W2115572397","https://openalex.org/W2117696986","https://openalex.org/W2123237149","https://openalex.org/W2145287260","https://openalex.org/W2152839228","https://openalex.org/W2160815625","https://openalex.org/W2162390675","https://openalex.org/W2163605009","https://openalex.org/W2170503676","https://openalex.org/W2271840356","https://openalex.org/W2272300165","https://openalex.org/W2276486856","https://openalex.org/W2294282016","https://openalex.org/W2400718301","https://openalex.org/W2405102949","https://openalex.org/W2520083297","https://openalex.org/W2584311934","https://openalex.org/W2584616277","https://openalex.org/W2585560244","https://openalex.org/W2585774018","https://openalex.org/W2606722458","https://openalex.org/W2625954420","https://openalex.org/W4212788319","https://openalex.org/W6637373629","https://openalex.org/W6674479107","https://openalex.org/W6678292227","https://openalex.org/W6683722107","https://openalex.org/W6684191040","https://openalex.org/W6693859313"],"related_works":["https://openalex.org/W4297942731","https://openalex.org/W2800704601","https://openalex.org/W3205838256","https://openalex.org/W2943610686","https://openalex.org/W1967938402","https://openalex.org/W229781084","https://openalex.org/W2386041993","https://openalex.org/W1608572506","https://openalex.org/W2160474882","https://openalex.org/W2218038495"],"abstract_inverted_index":{"With":[0],"the":[1,33,39,42,49,68,98,103,130,142,192,238],"recent":[2],"advancement":[3],"of":[4,38,48,172],"multilayer":[5],"convolutional":[6,80,86],"neural":[7],"networks":[8,13,162],"(CNNs)":[9],"and":[10,29,35,58,71,88,101,107,152,160,219,222,232],"fully":[11],"connected":[12],"(FCNs),":[14],"deep":[15,145],"learning":[16,146],"has":[17],"achieved":[18],"amazing":[19],"success":[20],"in":[21,25],"many":[22],"areas,":[23],"especially":[24],"visual":[26],"content":[27],"understanding":[28],"classification.":[30],"To":[31],"improve":[32],"performance":[34,171,221],"energy":[36,223,235],"efficiency":[37,236],"computation-demanding":[40],"CNN,":[41],"FPGA-based":[43],"acceleration":[44],"emerges":[45],"as":[46],"one":[47],"most":[50],"attractive":[51],"alternatives.":[52],"In":[53],"this":[54,94,190],"paper,":[55],"we":[56,76,96,118,138],"design":[57,119],"implement":[59],"Caffeine,":[60],"a":[61,78,113,135,169,181,228,244,247],"hardware/software":[62],"co-designed":[63],"library":[64],"to":[65,123,129,187,217],"efficiently":[66],"accelerate":[67],"entire":[69],"CNN":[70],"FCN":[72,90,203],"on":[73,93,112,163,180,202,227],"FPGAs.":[74],"First,":[75],"propose":[77],"uniformed":[79],"matrix-multiplication":[81],"representation":[82],"for":[83,243],"both":[84],"computation-bound":[85],"layers":[87,204],"communication-bound":[89],"layers.":[91],"Based":[92],"representation,":[95],"optimize":[97],"accelerator":[99],"microarchitecture":[100],"maximize":[102],"underlying":[104],"FPGA":[105,132,165,185,207,249],"computing":[106],"bandwidth":[108],"resource":[109],"utilization":[110],"based":[111],"revised":[114],"roofline":[115],"model.":[116],"Moreover,":[117],"an":[120],"automation":[121],"flow":[122],"directly":[124],"compile":[125],"highlevel":[126],"network":[127],"definitions":[128],"final":[131],"accelerator.":[133],"As":[134],"case":[136],"study,":[137],"integrate":[139],"Caffeine":[140,151,167],"into":[141],"industry-standard":[143],"software":[144],"framework":[147],"Caffe.":[148],"We":[149],"evaluate":[150],"its":[153],"integration":[154,214],"with":[155,212,246],"Caffe":[156,213,226],"by":[157],"implementing":[158],"VGG16":[159],"AlexNet":[161],"multiple":[164],"platforms.":[166],"achieves":[168,197],"peak":[170],"1460":[173],"giga":[174],"fixed":[175],"point":[176],"operations":[177],"per":[178],"second":[179],"medium-sized":[182],"Xilinx":[183],"KU060":[184],"board;":[186],"our":[188],"knowledge,":[189],"is":[191],"best":[193],"published":[194],"result.":[195],"It":[196],"more":[198],"than":[199],"100\u00d7":[200],"speedup":[201],"over":[205,225,237],"prior":[206],"accelerators.":[208],"An":[209],"end-to-end":[210],"evaluation":[211],"shows":[215],"up":[216],"29\u00d7":[218],"150\u00d7":[220],"gains":[224],"12-core":[229],"Xeon":[230],"server,":[231],"5.7\u00d7":[233],"better":[234],"GPU":[239],"implementation.":[240],"Performance":[241],"projections":[242],"system":[245],"high-end":[248],"(Virtex7":[250],"690t)":[251],"show":[252],"even":[253],"higher":[254],"gains.":[255]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":10},{"year":2024,"cited_by_count":39},{"year":2023,"cited_by_count":40},{"year":2022,"cited_by_count":37},{"year":2021,"cited_by_count":64},{"year":2020,"cited_by_count":44},{"year":2019,"cited_by_count":24},{"year":2018,"cited_by_count":3}],"updated_date":"2026-03-18T14:38:29.013473","created_date":"2025-10-10T00:00:00"}