{"id":"https://openalex.org/W4413267655","doi":"https://doi.org/10.1109/tcasai.2025.3592864","title":"Variation-Aware Demonstration and Optimization of Block Floating Point and Integer Neural Network Acceleration on RRAM Compute in-Memory Hardware","display_name":"Variation-Aware Demonstration and Optimization of Block Floating Point and Integer Neural Network Acceleration on RRAM Compute in-Memory Hardware","publication_year":2025,"publication_date":"2025-07-28","ids":{"openalex":"https://openalex.org/W4413267655","doi":"https://doi.org/10.1109/tcasai.2025.3592864"},"language":"en","primary_location":{"id":"doi:10.1109/tcasai.2025.3592864","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcasai.2025.3592864","pdf_url":null,"source":{"id":"https://openalex.org/S4404675360","display_name":"IEEE transactions on circuits and systems for artificial intelligence.","issn_l":"2996-6647","issn":["2996-6647"],"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Circuits and Systems for Artificial Intelligence","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/A5048307680","display_name":"Ahmed Shaban","orcid":"https://orcid.org/0000-0001-9528-1076"},"institutions":[{"id":"https://openalex.org/I68891433","display_name":"Indian Institute of Technology Delhi","ror":"https://ror.org/049tgcd06","country_code":"IN","type":"education","lineage":["https://openalex.org/I68891433"]}],"countries":["IN"],"is_corresponding":true,"raw_author_name":"Ahmed Shaban","raw_affiliation_strings":["Department of Electrical Engineering, Indian Institute of Technology, New Delhi, India","Department of Electrical Engineering, Indian Institute of Technology, New Delhi"],"affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, Indian Institute of Technology, New Delhi, India","institution_ids":["https://openalex.org/I68891433"]},{"raw_affiliation_string":"Department of Electrical Engineering, Indian Institute of Technology, New Delhi","institution_ids":["https://openalex.org/I68891433"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106329716","display_name":"Shreshtha Gothalyan","orcid":"https://orcid.org/0009-0002-4416-8721"},"institutions":[{"id":"https://openalex.org/I148366613","display_name":"National Yang Ming Chiao Tung University","ror":"https://ror.org/00se2k293","country_code":"TW","type":"education","lineage":["https://openalex.org/I148366613"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Shreshtha Gothalyan","raw_affiliation_strings":["Department of Electrical Engineering and Computer Science, National Yang Ming Chiao Tung University, Hsinchu, Taiwan"],"affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering and Computer Science, National Yang Ming Chiao Tung University, Hsinchu, Taiwan","institution_ids":["https://openalex.org/I148366613"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008011360","display_name":"Manan Suri","orcid":"https://orcid.org/0000-0003-1417-3570"},"institutions":[{"id":"https://openalex.org/I68891433","display_name":"Indian Institute of Technology Delhi","ror":"https://ror.org/049tgcd06","country_code":"IN","type":"education","lineage":["https://openalex.org/I68891433"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Manan Suri","raw_affiliation_strings":["Department of Electrical Engineering, Indian Institute of Technology, New Delhi, India","Department of Electrical Engineering, Indian Institute of Technology, Hauz Khas, New Delhi, India"],"affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, Indian Institute of Technology, New Delhi, India","institution_ids":["https://openalex.org/I68891433"]},{"raw_affiliation_string":"Department of Electrical Engineering, Indian Institute of Technology, Hauz Khas, New Delhi, India","institution_ids":["https://openalex.org/I68891433"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5024381546","display_name":"Tuo\u2010Hung Hou","orcid":"https://orcid.org/0000-0002-9686-7076"},"institutions":[{"id":"https://openalex.org/I148366613","display_name":"National Yang Ming Chiao Tung University","ror":"https://ror.org/00se2k293","country_code":"TW","type":"education","lineage":["https://openalex.org/I148366613"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Tuo-Hung Hou","raw_affiliation_strings":["Department of Electrical Engineering and Institute of Electronics, National Yang Ming Chiao Tung University, Hsinchu, Taiwan"],"affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering and Institute of Electronics, National Yang Ming Chiao Tung University, Hsinchu, Taiwan","institution_ids":["https://openalex.org/I148366613"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5048307680"],"corresponding_institution_ids":["https://openalex.org/I68891433"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.24570586,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"2","issue":"3","first_page":"185","last_page":"196"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":1.0,"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"}},"topics":[{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":1.0,"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":0.9994000196456909,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9945999979972839,"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/variation","display_name":"Variation (astronomy)","score":0.6993805170059204},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.6432452201843262},{"id":"https://openalex.org/keywords/integer","display_name":"Integer (computer science)","score":0.5892157554626465},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5649471879005432},{"id":"https://openalex.org/keywords/resistive-random-access-memory","display_name":"Resistive random-access memory","score":0.5560120940208435},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5339934229850769},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.5205629467964172},{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.5189268589019775},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.5185337662696838},{"id":"https://openalex.org/keywords/floating-point","display_name":"Floating point","score":0.4955497980117798},{"id":"https://openalex.org/keywords/point","display_name":"Point (geometry)","score":0.44678863883018494},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3440507650375366},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.23679980635643005},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.2064628005027771},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.20034077763557434},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.18573641777038574},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.1384669840335846},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.12617123126983643},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.08208075165748596},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.07306909561157227},{"id":"https://openalex.org/keywords/combinatorics","display_name":"Combinatorics","score":0.06189015507698059}],"concepts":[{"id":"https://openalex.org/C2778334786","wikidata":"https://www.wikidata.org/wiki/Q1586270","display_name":"Variation (astronomy)","level":2,"score":0.6993805170059204},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.6432452201843262},{"id":"https://openalex.org/C97137487","wikidata":"https://www.wikidata.org/wiki/Q729138","display_name":"Integer (computer science)","level":2,"score":0.5892157554626465},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5649471879005432},{"id":"https://openalex.org/C182019814","wikidata":"https://www.wikidata.org/wiki/Q1143830","display_name":"Resistive random-access memory","level":3,"score":0.5560120940208435},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5339934229850769},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.5205629467964172},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.5189268589019775},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.5185337662696838},{"id":"https://openalex.org/C84211073","wikidata":"https://www.wikidata.org/wiki/Q117879","display_name":"Floating point","level":2,"score":0.4955497980117798},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.44678863883018494},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3440507650375366},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.23679980635643005},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2064628005027771},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.20034077763557434},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.18573641777038574},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.1384669840335846},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.12617123126983643},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.08208075165748596},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.07306909561157227},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.06189015507698059},{"id":"https://openalex.org/C74650414","wikidata":"https://www.wikidata.org/wiki/Q11397","display_name":"Classical mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C44870925","wikidata":"https://www.wikidata.org/wiki/Q37547","display_name":"Astrophysics","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}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcasai.2025.3592864","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcasai.2025.3592864","pdf_url":null,"source":{"id":"https://openalex.org/S4404675360","display_name":"IEEE transactions on circuits and systems for artificial intelligence.","issn_l":"2996-6647","issn":["2996-6647"],"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Circuits and Systems for Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.4000000059604645,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":32,"referenced_works":["https://openalex.org/W2004823737","https://openalex.org/W2168972302","https://openalex.org/W2620112626","https://openalex.org/W2790511620","https://openalex.org/W2799406003","https://openalex.org/W2919115771","https://openalex.org/W2960778947","https://openalex.org/W2970608956","https://openalex.org/W2990591126","https://openalex.org/W3013080934","https://openalex.org/W3015982917","https://openalex.org/W3028754834","https://openalex.org/W3133754064","https://openalex.org/W3134195144","https://openalex.org/W3134526034","https://openalex.org/W3135906938","https://openalex.org/W3183406752","https://openalex.org/W3192589754","https://openalex.org/W4207063504","https://openalex.org/W4220958508","https://openalex.org/W4221038786","https://openalex.org/W4221101426","https://openalex.org/W4225580377","https://openalex.org/W4226111665","https://openalex.org/W4292121737","https://openalex.org/W4297812065","https://openalex.org/W4360605703","https://openalex.org/W4379659747","https://openalex.org/W4387092567","https://openalex.org/W4389459141","https://openalex.org/W4401568602","https://openalex.org/W4404520182"],"related_works":["https://openalex.org/W2054635671","https://openalex.org/W2545245183","https://openalex.org/W2017425642","https://openalex.org/W2350916061","https://openalex.org/W2952918855","https://openalex.org/W1970117475","https://openalex.org/W4396815615","https://openalex.org/W2146872326","https://openalex.org/W3158825072","https://openalex.org/W2586397364"],"abstract_inverted_index":{"In":[0,95],"this":[1],"work,":[2],"we":[3],"propose":[4],"a":[5,15,53,97,165],"device-circuit-system":[6],"co-design":[7],"approach":[8],"to":[9,113,128,163,193,263,288],"achieve":[10],"the":[11,219,223,246,250,258,290],"best":[12],"performance":[13,256],"from":[14,191],"given":[16],"variation-prone":[17],"resistive":[18],"random":[19],"access":[20],"memory":[21],"(RRAM)":[22],"compute-in-memory":[23],"(CIM)":[24],"hardware":[25,37,51,173,192,217],"for":[26,135,145,198],"deep":[27],"neural":[28],"network":[29,68,255],"(DNN)":[30],"acceleration":[31,44],"and":[32,60,86,110,138,206,211,297],"also":[33,84,161,179,286],"demonstrate":[34],"an":[35],"optimized":[36,157],"implementation":[38,127,150],"of":[39,56,196,222,249,292],"block":[40],"floating":[41],"point":[42],"(BFP)":[43],"exploiting":[45],"FP8":[46,175,224,264,272],"format":[47],"on":[48,74,119,152,172,257],"RRAM":[49,252,273],"CIM":[50],"using":[52,279],"1T-1R":[54,121,154],"array":[55,122,274],"single-level":[57],"cell":[58,62],"(SLC)":[59],"multilevel":[61],"(MLC)":[63],"devices.":[64],"Further,":[65],"bit-level":[66],"accurate":[67],"simulations":[69],"incorporating":[70],"device-to-device":[71],"(D2D)":[72],"variations":[73],"multi-bit":[75],"SLC":[76,202,215],"/":[77,81,203],"MLC":[78,167,204,281],"weight":[79,82,283],"mantissa":[80],"are":[83],"conducted":[85],"benchmarking":[87],"is":[88,108,160,178,234,261,285],"performed":[89,180],"with":[90,209],"integer":[91],"(INT4,":[92],"INT8)":[93],"formats.":[94],"particular,":[96],"word":[98],"line":[99],"read":[100],"voltage":[101],"(<inline-formula":[102],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[103,184,187,229,242,268],"xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math":[104,185,188,230,243,269],"notation=\"LaTeX\">$V_{\\boldsymbol{WLread}}$</tex-math></inline-formula>)":[105],"tuning":[106],"scheme":[107,159],"proposed":[109,162],"experimentally":[111],"analyzed":[112],"(a)":[114],"optimize":[115],"D2D":[116],"variation":[117,295],"(tested":[118,151],"144":[120],"bit":[123],"cells)":[124],"in":[125,148],"SLC-based":[126],"reduce":[129],"partial":[130],"MAC":[131],"(PMAC)":[132],"mapping":[133,205],"errors":[134],"robust":[136],"inference":[137,176,278],"b)":[139],"obtain":[140,164],"linearly":[141],"separated":[142],"conductance":[143],"states":[144,168],"linear":[146],"accumulation":[147],"MLC-based":[149],"200":[153],"bitcells).":[155],"An":[156],"write-verify":[158],"tight":[166],"distribution.":[169],"Extensive":[170],"analysis":[171],"obtained":[174],"results":[177],"by":[181],"measuring":[182],"16,384<inline-formula":[183],"notation=\"LaTeX\">$&lt;$</tex-math></inline-formula>PMACs<inline-formula":[186],"notation=\"LaTeX\">$&lt;$</tex-math></inline-formula>":[189],"1,31,072":[190],"identify":[194,298],"sources":[195],"error":[197],"different":[199],"word-line":[200],"parallelism,":[201],"their":[207],"trade-off":[208],"sparsity":[210],"energy":[212],"consumption.":[213,276],"For":[214],"based":[216],"implementation,":[218],"test":[220],"accuracy":[221],"trained":[225],"networks":[226],"at":[227,239,266],"<inline-formula":[228,241,267],"notation=\"LaTeX\">$\\sigma/\\mu":[231],"=$</tex-math></inline-formula>":[232],"0.5":[233],"1.2x":[235],"better":[236],"than":[237],"INT4":[238,251],"only":[240],"notation=\"LaTeX\">${\\boldsymbol\\sim}$</tex-math></inline-formula>":[244,270],"1.3x":[245],"area":[247,275],"consumption":[248],"array.":[253],"INT8":[254],"other":[259],"hand,":[260],"comparable":[262],"but":[265],"2x":[271],"Hardware":[277],"2-bit":[280],"mapped":[282],"mantissas":[284],"demonstrated":[287],"analyze":[289],"effect":[291],"individual":[293],"state":[294],"dependence":[296],"critical":[299],"variability":[300],"states.":[301]},"counts_by_year":[],"updated_date":"2025-12-23T23:11:35.936235","created_date":"2025-10-10T00:00:00"}