{"id":"https://openalex.org/W4385484619","doi":"https://doi.org/10.1109/ijcnn54540.2023.10191667","title":"Energy Efficient Memory-based Inference of LSTM by Exploiting FPGA Overlay","display_name":"Energy Efficient Memory-based Inference of LSTM by Exploiting FPGA Overlay","publication_year":2023,"publication_date":"2023-06-18","ids":{"openalex":"https://openalex.org/W4385484619","doi":"https://doi.org/10.1109/ijcnn54540.2023.10191667"},"language":"en","primary_location":{"id":"doi:10.1109/ijcnn54540.2023.10191667","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn54540.2023.10191667","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5002833632","display_name":"Krishnendu Guha","orcid":"https://orcid.org/0000-0003-1139-9582"},"institutions":[{"id":"https://openalex.org/I27577105","display_name":"University College Cork","ror":"https://ror.org/03265fv13","country_code":"IE","type":"education","lineage":["https://openalex.org/I27577105"]}],"countries":["IE"],"is_corresponding":false,"raw_author_name":"Krishnendu Guha","raw_affiliation_strings":["School of Computer Science and Information Technology, University College,Cork,Ireland","School of Computer Science and Information Technology, University College, Cork, Ireland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Information Technology, University College,Cork,Ireland","institution_ids":["https://openalex.org/I27577105"]},{"raw_affiliation_string":"School of Computer Science and Information Technology, University College, Cork, Ireland","institution_ids":["https://openalex.org/I27577105"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028132107","display_name":"Amit Ranjan Trivedi","orcid":"https://orcid.org/0000-0001-5436-7922"},"institutions":[{"id":"https://openalex.org/I39422238","display_name":"University of Illinois Chicago","ror":"https://ror.org/02mpq6x41","country_code":"US","type":"education","lineage":["https://openalex.org/I39422238"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Amit Ranjan Trivedi","raw_affiliation_strings":["University of Illinois at Chicago,Department of Electrical and Computer Engineering,USA","Department of Electrical and Computer Engineering, University of Illinois at Chicago, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Illinois at Chicago,Department of Electrical and Computer Engineering,USA","institution_ids":["https://openalex.org/I39422238"]},{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Illinois at Chicago, USA","institution_ids":["https://openalex.org/I39422238"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5039442844","display_name":"Swarup Bhunia","orcid":"https://orcid.org/0000-0001-6082-6961"},"institutions":[{"id":"https://openalex.org/I33213144","display_name":"University of Florida","ror":"https://ror.org/02y3ad647","country_code":"US","type":"education","lineage":["https://openalex.org/I33213144"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Swarup Bhunia","raw_affiliation_strings":["University of Florida,Department of Electrical and Computer Engineering,USA","Department of Electrical and Computer Engineering, University of Florida, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Florida,Department of Electrical and Computer Engineering,USA","institution_ids":["https://openalex.org/I33213144"]},{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Florida, USA","institution_ids":["https://openalex.org/I33213144"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9961000084877014,"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":0.9961000084877014,"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/T10320","display_name":"Neural Networks and Applications","score":0.992900013923645,"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/T14319","display_name":"Currency Recognition and Detection","score":0.9919000267982483,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8254102468490601},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.7660573720932007},{"id":"https://openalex.org/keywords/reconfigurability","display_name":"Reconfigurability","score":0.69060879945755},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5988417863845825},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.5358079671859741},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.5056126117706299},{"id":"https://openalex.org/keywords/overlay","display_name":"Overlay","score":0.500129222869873},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.45057278871536255},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.4404556453227997},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.37016981840133667},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2918814420700073}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8254102468490601},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.7660573720932007},{"id":"https://openalex.org/C2780149590","wikidata":"https://www.wikidata.org/wiki/Q7302742","display_name":"Reconfigurability","level":2,"score":0.69060879945755},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5988417863845825},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.5358079671859741},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.5056126117706299},{"id":"https://openalex.org/C136085584","wikidata":"https://www.wikidata.org/wiki/Q910289","display_name":"Overlay","level":2,"score":0.500129222869873},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.45057278871536255},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.4404556453227997},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.37016981840133667},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2918814420700073},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ijcnn54540.2023.10191667","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn54540.2023.10191667","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 International Joint Conference on Neural Networks (IJCNN)","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":0.8799999952316284}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":24,"referenced_works":["https://openalex.org/W1571140939","https://openalex.org/W2008692683","https://openalex.org/W2051247395","https://openalex.org/W2059314331","https://openalex.org/W2064675550","https://openalex.org/W2139535992","https://openalex.org/W2322493561","https://openalex.org/W2548438367","https://openalex.org/W2585720638","https://openalex.org/W2904773682","https://openalex.org/W2915106038","https://openalex.org/W2940345045","https://openalex.org/W2962820060","https://openalex.org/W3034833480","https://openalex.org/W3090581962","https://openalex.org/W3134705418","https://openalex.org/W3138893848","https://openalex.org/W3141650351","https://openalex.org/W3198121252","https://openalex.org/W4294691307","https://openalex.org/W6634338523","https://openalex.org/W6652653249","https://openalex.org/W6729264839","https://openalex.org/W6779777845"],"related_works":["https://openalex.org/W1544665014","https://openalex.org/W2748364266","https://openalex.org/W4243861219","https://openalex.org/W2556374054","https://openalex.org/W2069545207","https://openalex.org/W2092579166","https://openalex.org/W2969723039","https://openalex.org/W1977947048","https://openalex.org/W2914540329","https://openalex.org/W2159103767"],"abstract_inverted_index":{"The":[0],"fourth":[1],"industrial":[2,41],"revolution":[3],"(a.k.a.":[4],"Industry":[5,65,102],"4.0)":[6],"relies":[7],"on":[8,74,124,208],"intelligent":[9,34],"machines":[10,35],"that":[11],"are":[12,36,49,116,176],"fully":[13],"autonomous":[14],"and":[15,18,61,76,113,115,127,159,192,220],"can":[16],"diagnose":[17],"resolve":[19],"operational":[20],"issues":[21],"without":[22,200],"human":[23],"intervention.":[24],"Therefore,":[25,92],"embedded":[26,54,165],"computing":[27,55],"platforms":[28],"enabling":[29],"the":[30,39,93,136,163,180,209,229],"necessary":[31,181],"computations":[32],"for":[33,38,52,100,108],"critical":[37],"ongoing":[40],"revolution.":[42],"Especially":[43],"field":[44],"programmable":[45],"gate":[46],"arrays":[47],"(FPGAs)":[48],"highly":[50],"suited":[51],"such":[53,68],"due":[56],"to":[57,131,178],"their":[58],"high":[59],"performance":[60],"easy":[62],"reconfigurability.":[63],"Many":[64],"4.0":[66,103],"applications,":[67],"as":[69],"predictive":[70],"maintenance,":[71],"critically":[72],"depend":[73],"real-time":[75],"reliable":[77],"processing":[78],"of":[79,96,145,169,203],"time-series":[80],"data":[81],"using":[82],"recurrent":[83],"neural":[84],"network":[85],"models,":[86],"especially":[87],"long":[88],"short-term":[89],"memory":[90,166],"(LSTM).":[91],"FPGA-based":[94],"acceleration":[95],"LSTM":[97,106,146,198,232],"is":[98,226],"imperative":[99],"many":[101],"applications.":[104],"Existing":[105],"models":[107,199],"FPGAs":[109],"incur":[110],"significant":[111],"resources":[112],"power":[114,128,225],"not":[117],"energy":[118,194],"efficient.":[119],"Moreover,":[120],"prior":[121],"works":[122],"focusing":[123],"reducing":[125],"latency":[126,219],"mainly":[129],"adhere":[130],"model":[132],"pruning,":[133],"which":[134],"compromises":[135],"accuracy.":[137,204],"Comparatively,":[138],"we":[139,155,186],"propose":[140],"a":[141,214],"memory-based":[142],"energy-efficient":[143],"inference":[144],"by":[147],"exploiting":[148],"overlay":[149],"in":[150,162,218,224],"FPGA.":[151,171],"In":[152],"our":[153],"methodology,":[154,185],"pre-compute":[156],"predominant":[157],"operations":[158],"store":[160],"them":[161],"available":[164],"blocks":[167],"(EMBs)":[168],"an":[170],"On-demand,":[172],"these":[173],"pre-computed":[174],"results":[175],"accessed":[177],"minimize":[179],"workload.":[182],"Via":[183],"this":[184],"obtained":[187,227],"lower":[188,190],"latency,":[189],"power,":[191],"better":[193],"efficiency":[195],"than":[196],"state-of-the-art":[197],"any":[201],"loss":[202],"Specifically,":[205],"when":[206],"implemented":[207],"ZynQ":[210],"XCU104":[211],"evaluation":[212],"board,":[213],"3":[215],"x":[216,222],"reduction":[217,223],"5":[221],"then":[228],"reference":[230],"16-bit":[231],"model.":[233]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}