{"id":"https://openalex.org/W4291171966","doi":"https://doi.org/10.1109/tcad.2022.3198036","title":"A Novel Low-Power Compression Scheme for Systolic Array-Based Deep Learning Accelerators","display_name":"A Novel Low-Power Compression Scheme for Systolic Array-Based Deep Learning Accelerators","publication_year":2022,"publication_date":"2022-08-10","ids":{"openalex":"https://openalex.org/W4291171966","doi":"https://doi.org/10.1109/tcad.2022.3198036"},"language":"en","primary_location":{"id":"doi:10.1109/tcad.2022.3198036","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2022.3198036","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/A5070147631","display_name":"A. Arunachalam","orcid":"https://orcid.org/0000-0002-0750-0484"},"institutions":[{"id":"https://openalex.org/I162577319","display_name":"The University of Texas at Dallas","ror":"https://ror.org/049emcs32","country_code":"US","type":"education","lineage":["https://openalex.org/I162577319"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Ayush Arunachalam","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Texas at Dallas, Richardson, TX, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Texas at Dallas, Richardson, TX, USA","institution_ids":["https://openalex.org/I162577319"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066094401","display_name":"Shamik Kundu","orcid":"https://orcid.org/0000-0002-5992-8554"},"institutions":[{"id":"https://openalex.org/I162577319","display_name":"The University of Texas at Dallas","ror":"https://ror.org/049emcs32","country_code":"US","type":"education","lineage":["https://openalex.org/I162577319"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shamik Kundu","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Texas at Dallas, Richardson, TX, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Texas at Dallas, Richardson, TX, USA","institution_ids":["https://openalex.org/I162577319"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066464351","display_name":"Arnab Raha","orcid":"https://orcid.org/0000-0002-8848-1069"},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Arnab Raha","raw_affiliation_strings":["Intel Edge.AI, Intel Corporation, Santa Clara, CA, USA"],"affiliations":[{"raw_affiliation_string":"Intel Edge.AI, Intel Corporation, Santa Clara, CA, USA","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083178603","display_name":"Suvadeep Banerjee","orcid":"https://orcid.org/0000-0001-5188-1651"},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Suvadeep Banerjee","raw_affiliation_strings":["Intel Edge.AI, Intel Corporation, Santa Clara, CA, USA"],"affiliations":[{"raw_affiliation_string":"Intel Edge.AI, Intel Corporation, Santa Clara, CA, USA","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113724917","display_name":"Suriyaprakash Natarajan","orcid":"https://orcid.org/0000-0002-5499-4341"},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Suriyaprakash Natarajan","raw_affiliation_strings":["Intel Edge.AI, Intel Corporation, Santa Clara, CA, USA"],"affiliations":[{"raw_affiliation_string":"Intel Edge.AI, Intel Corporation, Santa Clara, CA, USA","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5066320524","display_name":"Kanad Basu","orcid":"https://orcid.org/0000-0002-6431-7512"},"institutions":[{"id":"https://openalex.org/I162577319","display_name":"The University of Texas at Dallas","ror":"https://ror.org/049emcs32","country_code":"US","type":"education","lineage":["https://openalex.org/I162577319"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kanad Basu","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Texas at Dallas, Richardson, TX, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Texas at Dallas, Richardson, TX, USA","institution_ids":["https://openalex.org/I162577319"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5070147631"],"corresponding_institution_ids":["https://openalex.org/I162577319"],"apc_list":null,"apc_paid":null,"fwci":0.5031,"has_fulltext":false,"cited_by_count":5,"citation_normalized_percentile":{"value":0.63231457,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":97},"biblio":{"volume":"42","issue":"4","first_page":"1085","last_page":"1098"},"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":0.9988999962806702,"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/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.998199999332428,"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/mnist-database","display_name":"MNIST database","score":0.7455496788024902},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7310090661048889},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6389294266700745},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.6244322061538696},{"id":"https://openalex.org/keywords/dram","display_name":"Dram","score":0.5280412435531616},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.5245345234870911},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.47754964232444763},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.47745636105537415},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.44069138169288635},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.43362003564834595},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.4186217188835144},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3808354139328003},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.3372393548488617},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.33593249320983887},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.329473078250885},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.2679952383041382},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.24209064245224},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.10291731357574463},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.08149746060371399}],"concepts":[{"id":"https://openalex.org/C190502265","wikidata":"https://www.wikidata.org/wiki/Q17069496","display_name":"MNIST database","level":3,"score":0.7455496788024902},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7310090661048889},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6389294266700745},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6244322061538696},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.5280412435531616},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.5245345234870911},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.47754964232444763},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.47745636105537415},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.44069138169288635},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.43362003564834595},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.4186217188835144},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3808354139328003},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3372393548488617},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.33593249320983887},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.329473078250885},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2679952383041382},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.24209064245224},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.10291731357574463},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.08149746060371399}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcad.2022.3198036","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2022.3198036","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":[],"awards":[{"id":"https://openalex.org/G7526799016","display_name":null,"funder_award_id":"2930.001","funder_id":"https://openalex.org/F4320306087","funder_display_name":"Semiconductor Research Corporation"}],"funders":[{"id":"https://openalex.org/F4320306087","display_name":"Semiconductor Research Corporation","ror":"https://ror.org/047z4n946"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":65,"referenced_works":["https://openalex.org/W1704989917","https://openalex.org/W1998305460","https://openalex.org/W1998917233","https://openalex.org/W2020217519","https://openalex.org/W2026005150","https://openalex.org/W2081858681","https://openalex.org/W2108481929","https://openalex.org/W2108598243","https://openalex.org/W2119144962","https://openalex.org/W2147267708","https://openalex.org/W2161651385","https://openalex.org/W2285660444","https://openalex.org/W2289252105","https://openalex.org/W2399535694","https://openalex.org/W2515531880","https://openalex.org/W2535375934","https://openalex.org/W2604319603","https://openalex.org/W2606722458","https://openalex.org/W2746351350","https://openalex.org/W2770370002","https://openalex.org/W2886851211","https://openalex.org/W2894740066","https://openalex.org/W2900227741","https://openalex.org/W2915589364","https://openalex.org/W2924515500","https://openalex.org/W2928560789","https://openalex.org/W2944349874","https://openalex.org/W2945065183","https://openalex.org/W2946682676","https://openalex.org/W2963122961","https://openalex.org/W2963363373","https://openalex.org/W2964108906","https://openalex.org/W2965862774","https://openalex.org/W2974514820","https://openalex.org/W2979535540","https://openalex.org/W2982479999","https://openalex.org/W2989762710","https://openalex.org/W2996011786","https://openalex.org/W2996874060","https://openalex.org/W2997768846","https://openalex.org/W3010745217","https://openalex.org/W3033428961","https://openalex.org/W3033788476","https://openalex.org/W3040024858","https://openalex.org/W3047984063","https://openalex.org/W3101923770","https://openalex.org/W3104849992","https://openalex.org/W3105131457","https://openalex.org/W3112402894","https://openalex.org/W3126291328","https://openalex.org/W3131039502","https://openalex.org/W3159119910","https://openalex.org/W3168100846","https://openalex.org/W3176282186","https://openalex.org/W3184376546","https://openalex.org/W3195381335","https://openalex.org/W4206835082","https://openalex.org/W4236853429","https://openalex.org/W4246554365","https://openalex.org/W4253012315","https://openalex.org/W6677580257","https://openalex.org/W6755166560","https://openalex.org/W6759263581","https://openalex.org/W6763207085","https://openalex.org/W6770165107"],"related_works":["https://openalex.org/W2618574054","https://openalex.org/W4385524141","https://openalex.org/W3018979822","https://openalex.org/W3026616975","https://openalex.org/W4288018014","https://openalex.org/W4297776111","https://openalex.org/W2989784533","https://openalex.org/W2996058201","https://openalex.org/W2946347869","https://openalex.org/W3127679336"],"abstract_inverted_index":{"The":[0,108],"proliferation":[1],"of":[2,14,32,42,58,60,62,85,115,165,179,191,210,237,274],"deep":[3,20],"learning":[4],"algorithms":[5],"has":[6,171,197],"catalyzed":[7],"their":[8,119],"utilization":[9,280],"to":[10,82,99,175,240],"solve":[11],"a":[12,126,201,234],"multitude":[13],"real-world":[15],"problems.":[16],"Algorithms":[17],"such":[18],"as":[19],"neural":[21],"networks":[22,142],"(DNNs)":[23],"are":[24],"compute-":[25],"and":[26,122,156,194,252],"power-intensive,":[27],"thereby":[28],"accentuating":[29],"the":[30,68,83,89,133,137,145,148,153,161,177,180,188,192,207,211,259],"development":[31],"hardware":[33,128],"platforms":[34],"like":[35],"DNN":[36,101,262],"inference":[37,40,131],"accelerators.":[38,53],"However,":[39],"execution":[41],"large":[43,55,157,231],"DNNs":[44,56,158],"in":[45,51,103,132,230,249,258,281],"resource-constrained":[46,104,282],"environments":[47],"induces":[48],"energy":[49,184],"bottlenecks":[50],"these":[52],"Since":[54],"consist":[57],"hundreds":[59],"millions":[61],"trained":[63,143,151,159],"parameters,":[64],"accessing":[65],"them":[66],"from":[67],"accelerator":[69],"memory":[70,250],"incurs":[71,270],"substantial":[72],"energy.":[73],"To":[74],"address":[75],"this":[76],"challenge,":[77],"we":[78],"propose":[79],"HardCompress,":[80,225],"which,":[81],"best":[84],"our":[86,265],"knowledge,":[87],"is":[88,218],"first":[90],"low-power":[91,127,267],"solution":[92,139],"that":[93,186,224],"uses":[94],"traditional":[95],"compression":[96,121,236],"strategies":[97],"pertaining":[98],"commercial":[100],"accelerators":[102],"IoT":[105],"edge":[106],"devices.":[107],"three-step":[109],"approach":[110],"involves":[111],"hardware-based":[112],"post-quantization":[113],"trimming":[114],"weights,":[116],"followed":[117],"by":[118,125],"dictionary-based":[120],"subsequent":[123],"decompression":[124,268],"engine":[129,269],"during":[130],"accelerator.":[134,263],"We":[135],"evaluate":[136],"proposed":[138,181,266],"on":[140,144,152,160],"lightweight":[141],"MNIST":[146],"dataset,":[147,155],"compact":[149],"model":[150,213],"CIFAR-10":[154],"ImageNet":[162],"dataset.":[163],"Performance":[164],"HardCompress":[166],"at":[167],"different":[168],"quantization":[169,257],"levels":[170],"been":[172,198],"analyzed.":[173],"Furthermore,":[174,264],"quantify":[176],"effectiveness":[178],"solution,":[182],"an":[183,271],"framework":[185,204],"contrasts":[187],"DRAM":[189],"energies":[190],"original":[193,212],"HardCompressed":[195,216],"models":[196],"developed.":[199],"Finally,":[200],"fault":[202,208],"injection":[203],"which":[205],"compares":[206],"resilience":[209],"with":[214],"its":[215],"counterpart":[217],"also":[219],"proposed.":[220],"Our":[221],"results":[222],"exhibit":[223],"without":[226],"any":[227],"performance":[228],"degradation":[229],"DNNs,":[232],"furnishes":[233],"maximum":[235],"99.27%,":[238],"equivalent":[239],"<inline-formula":[241],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[242],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">":[243],"<tex-math":[244],"notation=\"LaTeX\">$137\\times":[245],"$":[246],"</tex-math></inline-formula>":[247],"reduction":[248],"footprint":[251],"0.07":[253],"J":[254],"for":[255],"8-bit":[256],"systolic":[260],"array-based":[261],"area":[272],"overhead":[273],"only":[275],"0.02%;":[276],"thus,":[277],"enabling":[278],"HardCompress\u2019":[279],"environments.":[283]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2023,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}