{"id":"https://openalex.org/W4385584193","doi":"https://doi.org/10.1145/3611673","title":"Mitigating Memory Wall Effects in CNN Engines with On-the-Fly Weights Generation","display_name":"Mitigating Memory Wall Effects in CNN Engines with On-the-Fly Weights Generation","publication_year":2023,"publication_date":"2023-08-04","ids":{"openalex":"https://openalex.org/W4385584193","doi":"https://doi.org/10.1145/3611673"},"language":"en","primary_location":{"id":"doi:10.1145/3611673","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3611673","pdf_url":null,"source":{"id":"https://openalex.org/S105046310","display_name":"ACM Transactions on Design Automation of Electronic Systems","issn_l":"1084-4309","issn":["1084-4309","1557-7309"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Design Automation of Electronic 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/A5033442931","display_name":"Stylianos I. Venieris","orcid":"https://orcid.org/0000-0001-5181-6251"},"institutions":[{"id":"https://openalex.org/I4210117523","display_name":"Samsung (United Kingdom)","ror":"https://ror.org/01w6gjq94","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210117523"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Stylianos I. Venieris","raw_affiliation_strings":["Samsung AI Center, Cambridge, UK"],"raw_orcid":"https://orcid.org/0000-0001-5181-6251","affiliations":[{"raw_affiliation_string":"Samsung AI Center, Cambridge, UK","institution_ids":["https://openalex.org/I4210117523"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031116318","display_name":"Javier Fern\u00e1ndez-Marqu\u00e9s","orcid":null},"institutions":[{"id":"https://openalex.org/I4210117523","display_name":"Samsung (United Kingdom)","ror":"https://ror.org/01w6gjq94","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210117523"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Javier Fernandez-Marques","raw_affiliation_strings":["Samsung AI Center, Cambridge, UK"],"raw_orcid":"https://orcid.org/0000-0003-3747-6523","affiliations":[{"raw_affiliation_string":"Samsung AI Center, Cambridge, UK","institution_ids":["https://openalex.org/I4210117523"]}]},{"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/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]},{"id":"https://openalex.org/I4210117523","display_name":"Samsung (United Kingdom)","ror":"https://ror.org/01w6gjq94","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210117523"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Nicholas D. Lane","raw_affiliation_strings":["Samsung AI Center, Cambridge & University of Cambridge, UK"],"raw_orcid":"https://orcid.org/0000-0002-2728-8273","affiliations":[{"raw_affiliation_string":"Samsung AI Center, Cambridge & University of Cambridge, UK","institution_ids":["https://openalex.org/I241749","https://openalex.org/I4210117523"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.2931,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.49706412,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":96},"biblio":{"volume":"28","issue":"6","first_page":"1","last_page":"31"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":1.0,"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":1.0,"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":0.9998999834060669,"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.9962999820709229,"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":0.909174919128418},{"id":"https://openalex.org/keywords/control-reconfiguration","display_name":"Control reconfiguration","score":0.6774940490722656},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.5865718722343445},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.5852472186088562},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.5107069611549377},{"id":"https://openalex.org/keywords/inference-engine","display_name":"Inference engine","score":0.4964292645454407},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.4832204580307007},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4731111228466034},{"id":"https://openalex.org/keywords/flexibility","display_name":"Flexibility (engineering)","score":0.46898728609085083},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.45314842462539673},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.4481489062309265},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.44195663928985596},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.4292737543582916},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.37832608819007874},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3241681456565857}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.909174919128418},{"id":"https://openalex.org/C119701452","wikidata":"https://www.wikidata.org/wiki/Q5165881","display_name":"Control reconfiguration","level":2,"score":0.6774940490722656},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.5865718722343445},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.5852472186088562},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.5107069611549377},{"id":"https://openalex.org/C46743427","wikidata":"https://www.wikidata.org/wiki/Q1341685","display_name":"Inference engine","level":3,"score":0.4964292645454407},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.4832204580307007},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4731111228466034},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.46898728609085083},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.45314842462539673},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.4481489062309265},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.44195663928985596},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.4292737543582916},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.37832608819007874},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3241681456565857},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3611673","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3611673","pdf_url":null,"source":{"id":"https://openalex.org/S105046310","display_name":"ACM Transactions on Design Automation of Electronic Systems","issn_l":"1084-4309","issn":["1084-4309","1557-7309"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Design Automation of Electronic Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","score":0.8600000143051147,"id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":76,"referenced_works":["https://openalex.org/W569478347","https://openalex.org/W1967267438","https://openalex.org/W1995319790","https://openalex.org/W2094756095","https://openalex.org/W2116258181","https://openalex.org/W2118673147","https://openalex.org/W2141038589","https://openalex.org/W2194775991","https://openalex.org/W2289252105","https://openalex.org/W2296514843","https://openalex.org/W2412782625","https://openalex.org/W2517041198","https://openalex.org/W2533484770","https://openalex.org/W2541839172","https://openalex.org/W2560674852","https://openalex.org/W2565125333","https://openalex.org/W2574797063","https://openalex.org/W2593390416","https://openalex.org/W2603836393","https://openalex.org/W2605487586","https://openalex.org/W2608239929","https://openalex.org/W2612076670","https://openalex.org/W2616014673","https://openalex.org/W2725922920","https://openalex.org/W2727238169","https://openalex.org/W2740539392","https://openalex.org/W2751366252","https://openalex.org/W2793184281","https://openalex.org/W2803642548","https://openalex.org/W2808648285","https://openalex.org/W2810861508","https://openalex.org/W2896983500","https://openalex.org/W2898907473","https://openalex.org/W2902895686","https://openalex.org/W2904902077","https://openalex.org/W2905006747","https://openalex.org/W2905515056","https://openalex.org/W2917673601","https://openalex.org/W2927638312","https://openalex.org/W2950896714","https://openalex.org/W2963701245","https://openalex.org/W2979310060","https://openalex.org/W2979439447","https://openalex.org/W2979515806","https://openalex.org/W2981941217","https://openalex.org/W2996485093","https://openalex.org/W3003453708","https://openalex.org/W3003760014","https://openalex.org/W3007444108","https://openalex.org/W3016735325","https://openalex.org/W3017521908","https://openalex.org/W3033696290","https://openalex.org/W3034100311","https://openalex.org/W3036033155","https://openalex.org/W3104393472","https://openalex.org/W3119549454","https://openalex.org/W3129147932","https://openalex.org/W3167824842","https://openalex.org/W4205474951","https://openalex.org/W4205540825","https://openalex.org/W4214651231","https://openalex.org/W4230315356","https://openalex.org/W4232902075","https://openalex.org/W4233429846","https://openalex.org/W4237249260","https://openalex.org/W4241916159","https://openalex.org/W4245551996","https://openalex.org/W4248003330","https://openalex.org/W4249459551","https://openalex.org/W4252553205","https://openalex.org/W4255534610","https://openalex.org/W4285531802","https://openalex.org/W4286696412","https://openalex.org/W4295112409","https://openalex.org/W4307616509","https://openalex.org/W4394670654"],"related_works":["https://openalex.org/W2357657342","https://openalex.org/W2153432761","https://openalex.org/W1580144672","https://openalex.org/W2152623100","https://openalex.org/W2142042635","https://openalex.org/W4248634784","https://openalex.org/W2103296973","https://openalex.org/W1988127757","https://openalex.org/W2165832238","https://openalex.org/W4250547108"],"abstract_inverted_index":{"The":[0,156],"unprecedented":[1],"accuracy":[2],"of":[3,12,43,65,71,94,111,118,152,177,183,249,275],"convolutional":[4],"neural":[5],"networks":[6],"(CNNs)":[7],"across":[8],"a":[9,27,57,116,122,143,160,167,272],"broad":[10],"range":[11,274],"AI":[13],"tasks":[14],"has":[15,37],"led":[16],"to":[17,90,125,133,204,210,266],"their":[18],"widespread":[19],"deployment":[20,64],"in":[21,40,109,231],"mobile":[22],"and":[23,31,86,264],"embedded":[24],"settings.":[25],"In":[26,50,102],"pursuit":[28],"for":[29,115,259],"high-performance":[30],"energy-efficient":[32],"inference,":[33],"significant":[34],"research":[35],"effort":[36],"been":[38],"invested":[39],"the":[41,63,69,91,98,107,127,150,173,180,200,205,227,239,260],"design":[42,59,114,224],"field-programmable":[44],"gate":[45],"array":[46],"(FPGA)\u2013based":[47],"CNN":[48,112,145,154,162,278],"accelerators.":[49,279],"this":[51,75,103],"context,":[52],"single":[53],"computation":[54],"engines":[55],"constitute":[56],"popular":[58],"approach":[60],"that":[61,120,148,165,171,198,225,238,245],"enables":[62,172],"diverse":[66,273],"models":[67,119],"without":[68],"overhead":[70],"fabric":[72],"reconfiguration.":[73],"Nevertheless,":[74],"flexibility":[76],"often":[77],"comes":[78],"with":[79,193],"significantly":[80],"degraded":[81],"performance":[82,251,269],"on":[83,97,186],"memory-bound":[84,187],"layers":[85,96],"resource":[87],"underutilisation":[88],"due":[89],"suboptimal":[92],"mapping":[93],"certain":[95],"engine\u2019s":[99],"fixed":[100],"configuration.":[101],"work,":[104],"we":[105,216],"investigate":[106],"implications":[108],"terms":[110],"engine":[113],"class":[117],"introduce":[121,217],"pre-convolution":[123],"stage":[124],"decompress":[126],"weights":[128,168,201],"at":[129],"runtime.":[130],"We":[131,189],"refer":[132],"these":[134],"approaches":[135],"as":[136],"on-the-fly":[137,175],".":[138],"This":[139],"article":[140],"presents":[141],"unzipFPGA,":[142],"novel":[144,161],"inference":[146],"system":[147],"counteracts":[149],"limitations":[151],"existing":[153],"engines.":[155],"proposed":[157,240],"framework":[158,241],"comprises":[159],"hardware":[163,243],"architecture":[164],"introduces":[166],"generator":[169],"module":[170],"on-chip":[174],"generation":[176,202],"weights,":[178],"alleviating":[179],"negative":[181],"impact":[182],"limited":[184],"bandwidth":[185],"layers.":[188,234],"further":[190],"enhance":[191],"unzipFPGA":[192],"an":[194,211,218,247],"automated":[195],"hardware-aware":[196],"methodology":[197],"tailors":[199],"mechanism":[203],"target":[206],"CNN-device":[207],"pair,":[208],"leading":[209],"improved":[212],"accuracy\u2013performance":[213],"balance.":[214],"Finally,":[215],"input":[219],"selective":[220],"processing":[221],"element":[222],"(PE)":[223],"balances":[226],"load":[228],"between":[229],"PEs":[230],"suboptimally":[232],"mapped":[233],"Quantitative":[235],"evaluation":[236],"shows":[237],"yields":[242],"designs":[244,258],"achieve":[246],"average":[248],"2.57\u00d7":[250],"efficiency":[252],"gain":[253],"over":[254,271],"highly":[255],"optimised":[256],"GPU":[257],"same":[261],"power":[262],"constraints":[263],"up":[265],"3.94\u00d7":[267],"higher":[268],"density":[270],"state-of-the-art":[276],"FPGA-based":[277]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":2}],"updated_date":"2026-07-22T07:51:19.307946","created_date":"2025-10-10T00:00:00"}