{"id":"https://openalex.org/W2916143956","doi":"https://doi.org/10.1145/3289602.3293967","title":"An FPGA-based Fine Tuning Accelerator for a Sparse CNN","display_name":"An FPGA-based Fine Tuning Accelerator for a Sparse CNN","publication_year":2019,"publication_date":"2019-02-20","ids":{"openalex":"https://openalex.org/W2916143956","doi":"https://doi.org/10.1145/3289602.3293967","mag":"2916143956"},"language":"en","primary_location":{"id":"doi:10.1145/3289602.3293967","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3289602.3293967","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2019 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays","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/A5070734898","display_name":"Hiroki Nakahara","orcid":"https://orcid.org/0000-0002-5701-7466"},"institutions":[{"id":"https://openalex.org/I114531698","display_name":"Tokyo Institute of Technology","ror":"https://ror.org/0112mx960","country_code":"JP","type":"education","lineage":["https://openalex.org/I114531698"]}],"countries":["JP"],"is_corresponding":true,"raw_author_name":"Hiroki Nakahara","raw_affiliation_strings":["Tokyo Institute of Technology, Meguro, Japan"],"affiliations":[{"raw_affiliation_string":"Tokyo Institute of Technology, Meguro, Japan","institution_ids":["https://openalex.org/I114531698"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016922897","display_name":"Akira Jinguji","orcid":"https://orcid.org/0000-0001-5691-3472"},"institutions":[{"id":"https://openalex.org/I114531698","display_name":"Tokyo Institute of Technology","ror":"https://ror.org/0112mx960","country_code":"JP","type":"education","lineage":["https://openalex.org/I114531698"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Akira Jinguji","raw_affiliation_strings":["Tokyo Institute of Technology, Meguro, Japan"],"affiliations":[{"raw_affiliation_string":"Tokyo Institute of Technology, Meguro, Japan","institution_ids":["https://openalex.org/I114531698"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043084588","display_name":"Masayuki Shimoda","orcid":"https://orcid.org/0000-0003-4627-0957"},"institutions":[{"id":"https://openalex.org/I114531698","display_name":"Tokyo Institute of Technology","ror":"https://ror.org/0112mx960","country_code":"JP","type":"education","lineage":["https://openalex.org/I114531698"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Masayuki Shimoda","raw_affiliation_strings":["Tokyo Institute of Technology, Meguro, Japan"],"affiliations":[{"raw_affiliation_string":"Tokyo Institute of Technology, Meguro, Japan","institution_ids":["https://openalex.org/I114531698"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101424734","display_name":"Shimpei Sato","orcid":"https://orcid.org/0000-0003-0292-1391"},"institutions":[{"id":"https://openalex.org/I114531698","display_name":"Tokyo Institute of Technology","ror":"https://ror.org/0112mx960","country_code":"JP","type":"education","lineage":["https://openalex.org/I114531698"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Shimpei Sato","raw_affiliation_strings":["Tokyo Institute of Technology, Meguro, Japan"],"affiliations":[{"raw_affiliation_string":"Tokyo Institute of Technology, Meguro, Japan","institution_ids":["https://openalex.org/I114531698"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5070734898"],"corresponding_institution_ids":["https://openalex.org/I114531698"],"apc_list":null,"apc_paid":null,"fwci":0.8098,"has_fulltext":false,"cited_by_count":8,"citation_normalized_percentile":{"value":0.76254293,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"186","last_page":"186"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9991000294685364,"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":0.9991000294685364,"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/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9902999997138977,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.9894000291824341,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8367221355438232},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.7651205062866211},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.7605297565460205},{"id":"https://openalex.org/keywords/pooling","display_name":"Pooling","score":0.636976957321167},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.554902970790863},{"id":"https://openalex.org/keywords/dram","display_name":"Dram","score":0.49217572808265686},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.4885495603084564},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.44511163234710693},{"id":"https://openalex.org/keywords/virtex","display_name":"Virtex","score":0.4155694842338562},{"id":"https://openalex.org/keywords/sparse-matrix","display_name":"Sparse matrix","score":0.4129795432090759},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.3388001322746277},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.30719560384750366}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8367221355438232},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.7651205062866211},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.7605297565460205},{"id":"https://openalex.org/C70437156","wikidata":"https://www.wikidata.org/wiki/Q7228652","display_name":"Pooling","level":2,"score":0.636976957321167},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.554902970790863},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.49217572808265686},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.4885495603084564},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.44511163234710693},{"id":"https://openalex.org/C2777674469","wikidata":"https://www.wikidata.org/wiki/Q20741011","display_name":"Virtex","level":3,"score":0.4155694842338562},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.4129795432090759},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.3388001322746277},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.30719560384750366},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3289602.3293967","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3289602.3293967","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2019 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.8999999761581421}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2953234277","https://openalex.org/W2626256601","https://openalex.org/W147410782","https://openalex.org/W2900413183","https://openalex.org/W4390975304","https://openalex.org/W3120961607","https://openalex.org/W3022252430","https://openalex.org/W2347980442","https://openalex.org/W2137627158","https://openalex.org/W1593523093"],"abstract_inverted_index":{"Fine-tuning":[0],"learns":[1],"abundant":[2],"feature":[3],"expression":[4],"for":[5,41,65,87],"a":[6,14,23,42,66,73,78,129],"wide":[7,24],"range":[8,25],"of":[9,26,52,106],"natural":[10],"images":[11],"by":[12,121],"using":[13],"pre-trained":[15,79],"CNN":[16,81],"model.":[17],"It":[18],"can":[19,149],"be":[20],"applied":[21],"to":[22,82,112,167],"the":[27,84,88,104,107,114,141,147,168],"neural":[28,45],"network":[29,46],"(NN)based":[30],"computer":[31],"vision":[32],"problems.":[33],"This":[34],"paper":[35,71],"proposes":[36],"an":[37],"FPGA-based":[38],"fine-tuning":[39,74,93],"accelerator":[40,145],"sparse":[43,53,67,80,143],"convolutional":[44,54],"(CNN).":[47],"The":[48],"proposed":[49,142],"architecture":[50],"consists":[51],"units":[55,58],"and":[56,100,117,159],"pooling":[57],"with":[59],"distributed":[60],"stacks":[61],"those":[62],"are":[63],"suitable":[64],"CNN.":[68],"Additionally,":[69],"this":[70],"presents":[72],"scheme,":[75],"which":[76],"loads":[77],"reduce":[83,118],"memory":[85],"size":[86],"training":[89,115],"step.":[90],"Thus,":[91],"our":[92],"scheme":[94],"stores":[95],"all":[96],"parameters":[97],"on":[98,128,146],"BRAMs":[99],"Ultra":[101],"RAMs":[102],"in":[103],"case":[105],"Xilinx":[108,130],"Virtex":[109,131],"UltraScale+":[110,132],"FPGA":[111,148],"accelerate":[113],"computation":[116],"power":[119,157],"consumption":[120],"eliminating":[122],"energy-costly":[123],"DRAM":[124],"accesses.":[125],"We":[126],"implemented":[127],"VC1525":[133],"acceleration":[134],"development":[135],"kit.":[136],"Experimental":[137],"results":[138],"show":[139],"that":[140],"finetuning":[144],"achieve":[150],"four":[151],"times":[152,155,161],"faster,":[153],"2.9":[154],"lower":[156],"consumption,":[158],"11.6":[160],"better":[162],"performance":[163],"per":[164],"power,":[165],"compared":[166],"existing":[169],"NVidia":[170],"GTX1080Ti":[171],"GPU.":[172]},"counts_by_year":[{"year":2021,"cited_by_count":1},{"year":2020,"cited_by_count":5},{"year":2019,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}