{"id":"https://openalex.org/W3191618777","doi":"https://doi.org/10.1109/islped52811.2021.9502486","title":"Efficacy of Pruning in Ultra-Low Precision DNNs","display_name":"Efficacy of Pruning in Ultra-Low Precision DNNs","publication_year":2021,"publication_date":"2021-07-26","ids":{"openalex":"https://openalex.org/W3191618777","doi":"https://doi.org/10.1109/islped52811.2021.9502486","mag":"3191618777"},"language":"en","primary_location":{"id":"doi:10.1109/islped52811.2021.9502486","is_oa":false,"landing_page_url":"https://doi.org/10.1109/islped52811.2021.9502486","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE/ACM International Symposium on Low Power Electronics and Design (ISLPED)","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/A5021284824","display_name":"Sanchari Sen","orcid":"https://orcid.org/0000-0003-0080-2882"},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]},{"id":"https://openalex.org/I4210114115","display_name":"IBM Research - Thomas J. Watson Research Center","ror":"https://ror.org/0265w5591","country_code":"US","type":"facility","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210114115"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Sanchari Sen","raw_affiliation_strings":["IBM T.J. Watson Research Center, Yorktown Heights, NY","School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN"],"affiliations":[{"raw_affiliation_string":"IBM T.J. Watson Research Center, Yorktown Heights, NY","institution_ids":["https://openalex.org/I4210114115"]},{"raw_affiliation_string":"School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN","institution_ids":["https://openalex.org/I219193219"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010094713","display_name":"Swagath Venkataramani","orcid":"https://orcid.org/0000-0002-0470-6364"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Swagath Venkataramani","raw_affiliation_strings":["Yorktown Heights, IBM T. J. Watson Research Center, NY"],"affiliations":[{"raw_affiliation_string":"Yorktown Heights, IBM T. J. Watson Research Center, NY","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5065766721","display_name":"Anand Raghunathan","orcid":"https://orcid.org/0000-0002-4624-564X"},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]},{"id":"https://openalex.org/I4210114115","display_name":"IBM Research - Thomas J. Watson Research Center","ror":"https://ror.org/0265w5591","country_code":"US","type":"facility","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210114115"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Anand Raghunathan","raw_affiliation_strings":["IBM T.J. Watson Research Center, Yorktown Heights, NY","School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN"],"affiliations":[{"raw_affiliation_string":"IBM T.J. Watson Research Center, Yorktown Heights, NY","institution_ids":["https://openalex.org/I4210114115"]},{"raw_affiliation_string":"School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN","institution_ids":["https://openalex.org/I219193219"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5021284824"],"corresponding_institution_ids":["https://openalex.org/I219193219","https://openalex.org/I4210114115"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.09813632,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"abs 1805 6085","issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12169","display_name":"Non-Destructive Testing Techniques","score":0.9980999827384949,"subfield":{"id":"https://openalex.org/subfields/2210","display_name":"Mechanical 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/T12169","display_name":"Non-Destructive Testing Techniques","score":0.9980999827384949,"subfield":{"id":"https://openalex.org/subfields/2210","display_name":"Mechanical 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/T11032","display_name":"VLSI and Analog Circuit Testing","score":0.9976999759674072,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T14117","display_name":"Integrated Circuits and Semiconductor Failure Analysis","score":0.9976000189781189,"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":0.7583150863647461},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.632999837398529},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.6172659397125244},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.6152381896972656},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.5152750015258789},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.476018488407135},{"id":"https://openalex.org/keywords/search-engine-indexing","display_name":"Search engine indexing","score":0.4588431119918823},{"id":"https://openalex.org/keywords/compressed-sensing","display_name":"Compressed sensing","score":0.4518078565597534},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.40420034527778625},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4025721549987793},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3620021641254425}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7583150863647461},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.632999837398529},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.6172659397125244},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.6152381896972656},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.5152750015258789},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.476018488407135},{"id":"https://openalex.org/C75165309","wikidata":"https://www.wikidata.org/wiki/Q2258979","display_name":"Search engine indexing","level":2,"score":0.4588431119918823},{"id":"https://openalex.org/C124851039","wikidata":"https://www.wikidata.org/wiki/Q2665459","display_name":"Compressed sensing","level":2,"score":0.4518078565597534},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.40420034527778625},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4025721549987793},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3620021641254425},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","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/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/islped52811.2021.9502486","is_oa":false,"landing_page_url":"https://doi.org/10.1109/islped52811.2021.9502486","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE/ACM International Symposium on Low Power Electronics and Design (ISLPED)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":29,"referenced_works":["https://openalex.org/W1506342804","https://openalex.org/W1982970530","https://openalex.org/W1998917233","https://openalex.org/W2101900253","https://openalex.org/W2164586147","https://openalex.org/W2285660444","https://openalex.org/W2604319603","https://openalex.org/W2606722458","https://openalex.org/W2612445135","https://openalex.org/W2625457103","https://openalex.org/W2751477244","https://openalex.org/W2764043458","https://openalex.org/W2786771851","https://openalex.org/W2889797931","https://openalex.org/W2898985762","https://openalex.org/W2945146780","https://openalex.org/W2951963285","https://openalex.org/W2962965870","https://openalex.org/W2963480671","https://openalex.org/W2964081807","https://openalex.org/W2964233199","https://openalex.org/W2964299589","https://openalex.org/W2984140583","https://openalex.org/W4297775537","https://openalex.org/W6726275242","https://openalex.org/W6737664043","https://openalex.org/W6743755670","https://openalex.org/W6745148473","https://openalex.org/W6748224102"],"related_works":["https://openalex.org/W2158224665","https://openalex.org/W2379589510","https://openalex.org/W3024364549","https://openalex.org/W4300044672","https://openalex.org/W2810730439","https://openalex.org/W4206019083","https://openalex.org/W1881631164","https://openalex.org/W2054476758","https://openalex.org/W1976265003","https://openalex.org/W2370378377"],"abstract_inverted_index":{"Quantization,":[0],"or":[1,11],"reducing":[2],"the":[3,22,55,63,71,81,89,103,114,121,163,192,209,218,242,248],"precision":[4,65],"of":[5,24,45,57,73,105,116,124,211],"variables":[6],"and":[7,9,14,41,78,134,139,171,194,237],"operations,":[8],"pruning,":[10],"removing":[12],"neurons":[13],"connections":[15],"are":[16],"two":[17,125],"popular":[18,126],"approaches":[19],"for":[20,60,88],"improving":[21],"efficiency":[23],"DNNs.":[25],"These":[26],"directions":[27],"have":[28],"been":[29],"pursued":[30],"largely":[31],"separately.":[32],"In":[33],"this":[34,155,176],"work,":[35],"we":[36,53,157,178],"investigate":[37],"combining":[38],"DNN":[39],"quantization":[40],"pruning":[42,58],"as":[43],"each":[44],"them":[46],"is":[47],"pushed":[48],"to":[49,76,110,147,207,257],"its":[50],"limits.":[51],"Specifically,":[52],"explore":[54],"efficacy":[56],"DNNs":[59],"inference":[61],"in":[62,150,227],"ultralow":[64],"(sub-8":[66],"bit)":[67],"regime.":[68],"Pruning":[69],"requires":[70],"use":[72],"sparse":[74,94,127,165,188],"formats":[75,128,189],"store":[77],"access":[79],"only":[80],"non-zero":[82,107],"values.":[83],"We":[84,96,118,197],"provide":[85],"analytical":[86],"expressions":[87],"storage":[90],"required":[91],"by":[92,221,252],"these":[93],"formats.":[95],"demonstrate":[97,140],"that":[98,141,162,184],"with":[99],"decreasing":[100],"weight":[101],"precision,":[102],"overhead":[104],"indexing":[106],"locations":[108],"starts":[109],"dominate,":[111],"greatly":[112],"diminishing":[113],"benefits":[115],"pruning.":[117],"specifically":[119],"analyze":[120],"compression":[122,182,223,245,250],"ratios":[123],"\u2013":[129,138],"Compressed":[130],"Sparse":[131],"Column":[132],"(CSC)":[133],"Sparsity":[135,204],"Map":[136,205],"(Smap)":[137],"they":[142],"drop":[143],"significantly,":[144],"even":[145],"degrading":[146],"<1":[148],"(inflation)":[149],"some":[151],"cases.":[152],"To":[153],"address":[154],"challenge,":[156],"make":[158],"a":[159,180,200],"key":[160],"observation":[161],"best-performing":[164],"format":[166],"varies":[167],"across":[168],"different":[169],"precisions":[170],"sparsity":[172],"levels.":[173],"Based":[174],"on":[175],"observation,":[177],"propose":[179,199],"hybrid":[181,212,244],"scheme":[183,216,220,246],"dynamically":[185],"chooses":[186],"between":[187],"at":[190],"both":[191],"network":[193],"layer-level":[195],"granularities.":[196],"further":[198],"new":[201],"scheme,":[202],"compressed":[203],"(cSmap),":[206],"enhance":[208],"performance":[210],"compression.":[213],"The":[214],"cSmap":[215],"improves":[217,247],"Smap":[219],"applying":[222],"methods":[224],"widely":[225],"used":[226],"manufacturing":[228],"test.":[229],"Across":[230],"6":[231],"state-of-the-art":[232],"Convolutional":[233],"Neural":[234,239],"Networks":[235,240],"(CNNs)":[236],"Recurrent":[238],"(RNNs),":[241],"proposed":[243],"average":[249],"ratio":[251],"18.3%":[253],"-":[254],"34.7%":[255],"compared":[256],"previous":[258],"approaches.":[259]},"counts_by_year":[],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}