{"id":"https://openalex.org/W2534720278","doi":"https://doi.org/10.1145/2966986.2967028","title":"Efficient memory compression in deep neural networks using coarse-grain sparsification for speech applications","display_name":"Efficient memory compression in deep neural networks using coarse-grain sparsification for speech applications","publication_year":2016,"publication_date":"2016-10-18","ids":{"openalex":"https://openalex.org/W2534720278","doi":"https://doi.org/10.1145/2966986.2967028","mag":"2534720278"},"language":"en","primary_location":{"id":"doi:10.1145/2966986.2967028","is_oa":false,"landing_page_url":"https://doi.org/10.1145/2966986.2967028","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 35th International Conference on Computer-Aided Design","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/A5085748811","display_name":"Deepak Kadetotad","orcid":"https://orcid.org/0000-0003-0924-7213"},"institutions":[{"id":"https://openalex.org/I55732556","display_name":"Arizona State University","ror":"https://ror.org/03efmqc40","country_code":"US","type":"education","lineage":["https://openalex.org/I55732556"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Deepak Kadetotad","raw_affiliation_strings":["Arizona State University, Tempe, AZ","School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, 85281, United States of America"],"affiliations":[{"raw_affiliation_string":"Arizona State University, Tempe, AZ","institution_ids":["https://openalex.org/I55732556"]},{"raw_affiliation_string":"School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, 85281, United States of America","institution_ids":["https://openalex.org/I55732556"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103717256","display_name":"Sairam Arunachalam","orcid":null},"institutions":[{"id":"https://openalex.org/I55732556","display_name":"Arizona State University","ror":"https://ror.org/03efmqc40","country_code":"US","type":"education","lineage":["https://openalex.org/I55732556"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sairam Arunachalam","raw_affiliation_strings":["Arizona State University, Tempe, AZ","School of Computing, Informatics, Decision Systems Engineering, Arizona State University, Tempe, 85281, United States of America"],"affiliations":[{"raw_affiliation_string":"Arizona State University, Tempe, AZ","institution_ids":["https://openalex.org/I55732556"]},{"raw_affiliation_string":"School of Computing, Informatics, Decision Systems Engineering, Arizona State University, Tempe, 85281, United States of America","institution_ids":["https://openalex.org/I55732556"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025336372","display_name":"Chaitali Chakrabarti","orcid":"https://orcid.org/0000-0002-9859-7778"},"institutions":[{"id":"https://openalex.org/I55732556","display_name":"Arizona State University","ror":"https://ror.org/03efmqc40","country_code":"US","type":"education","lineage":["https://openalex.org/I55732556"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chaitali Chakrabarti","raw_affiliation_strings":["Arizona State University, Tempe, AZ","School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, 85281, United States of America"],"affiliations":[{"raw_affiliation_string":"Arizona State University, Tempe, AZ","institution_ids":["https://openalex.org/I55732556"]},{"raw_affiliation_string":"School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, 85281, United States of America","institution_ids":["https://openalex.org/I55732556"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5007690955","display_name":"Jae-sun Seo","orcid":"https://orcid.org/0000-0002-4551-7789"},"institutions":[{"id":"https://openalex.org/I55732556","display_name":"Arizona State University","ror":"https://ror.org/03efmqc40","country_code":"US","type":"education","lineage":["https://openalex.org/I55732556"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jae-sun Seo","raw_affiliation_strings":["Arizona State University, Tempe, AZ","School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, 85281, United States of America"],"affiliations":[{"raw_affiliation_string":"Arizona State University, Tempe, AZ","institution_ids":["https://openalex.org/I55732556"]},{"raw_affiliation_string":"School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, 85281, United States of America","institution_ids":["https://openalex.org/I55732556"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5085748811"],"corresponding_institution_ids":["https://openalex.org/I55732556"],"apc_list":null,"apc_paid":null,"fwci":9.4265,"has_fulltext":false,"cited_by_count":34,"citation_normalized_percentile":{"value":0.97982251,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998000264167786,"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"}},"topics":[{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998000264167786,"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/T10860","display_name":"Speech and Audio Processing","score":0.9979000091552734,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T11309","display_name":"Music and Audio Processing","score":0.9976999759674072,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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":0.8309171199798584},{"id":"https://openalex.org/keywords/memory-footprint","display_name":"Memory footprint","score":0.6038472056388855},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5706201791763306},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.44154202938079834},{"id":"https://openalex.org/keywords/high-memory","display_name":"High memory","score":0.4306142330169678},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.4303876757621765},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.4253905415534973},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3421659469604492},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.3344813883304596},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3034396767616272},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.17212477326393127}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8309171199798584},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.6038472056388855},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5706201791763306},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.44154202938079834},{"id":"https://openalex.org/C2781357197","wikidata":"https://www.wikidata.org/wiki/Q5757597","display_name":"High memory","level":2,"score":0.4306142330169678},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.4303876757621765},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.4253905415534973},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3421659469604492},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3344813883304596},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3034396767616272},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.17212477326393127},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/2966986.2967028","is_oa":false,"landing_page_url":"https://doi.org/10.1145/2966986.2967028","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 35th International Conference on Computer-Aided Design","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","score":0.44999998807907104,"display_name":"Peace, Justice and strong institutions"}],"awards":[{"id":"https://openalex.org/G2320677022","display_name":null,"funder_award_id":"1535669","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320332195","display_name":"Samsung","ror":"https://ror.org/04w3jy968"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W4919037","https://openalex.org/W1524333225","https://openalex.org/W1643320849","https://openalex.org/W1661756259","https://openalex.org/W1902934009","https://openalex.org/W1984541135","https://openalex.org/W1998917233","https://openalex.org/W1999121090","https://openalex.org/W2001905757","https://openalex.org/W2005687631","https://openalex.org/W2038332023","https://openalex.org/W2056986588","https://openalex.org/W2084910356","https://openalex.org/W2091432990","https://openalex.org/W2119144962","https://openalex.org/W2143612262","https://openalex.org/W2155653793","https://openalex.org/W2183733815","https://openalex.org/W2295119550","https://openalex.org/W2407023693","https://openalex.org/W2618530766","https://openalex.org/W2963114950","https://openalex.org/W2963374099","https://openalex.org/W2964299589","https://openalex.org/W6600213771","https://openalex.org/W6631362777","https://openalex.org/W6638783484"],"related_works":["https://openalex.org/W4390846322","https://openalex.org/W2798332276","https://openalex.org/W4200309432","https://openalex.org/W4387677131","https://openalex.org/W4286229642","https://openalex.org/W2748428003","https://openalex.org/W4226138706","https://openalex.org/W3217186257","https://openalex.org/W2982185430","https://openalex.org/W2929170389"],"abstract_inverted_index":{"Recent":[0],"breakthroughs":[1],"in":[2,14,34,82,170,185,201],"deep":[3,20],"neural":[4,21,64],"networks":[5,22,27,65],"have":[6],"led":[7],"to":[8,46,60,103,122,159],"the":[9,48,98,119,132,140,150,214,221],"proliferation":[10],"of":[11,32,85,139],"its":[12],"use":[13],"image":[15],"and":[16,71,108,128,143,177,195,203,220,234],"speech":[17,129,222],"applications.":[18],"Conventional":[19],"(DNNs)":[23],"are":[24,135],"fully-connected":[25,161],"multi-layer":[26],"with":[28,66,137,145,163],"hundreds":[29],"or":[30],"thousands":[31],"neurons":[33],"each":[35],"layer.":[36],"Such":[37],"a":[38,41,57,188,192],"network":[39,217,224],"requires":[40],"very":[42],"large":[43,83],"weight":[44,105,148,151],"memory":[45,69,106,152],"store":[47],"connectivity":[49],"between":[50],"neurons.":[51],"In":[52,209],"this":[53,77,183,228],"paper,":[54],"we":[55],"propose":[56],"hardware-centric":[58],"methodology":[59],"design":[61,124],"low":[62,205],"power":[63,206],"significantly":[67],"smaller":[68],"footprint":[70],"computation":[72,110],"resource":[73],"requirements.":[74],"We":[75,117],"achieve":[76],"by":[78,156],"judiciously":[79],"dropping":[80],"connections":[81],"blocks":[84],"weights.":[86],"The":[87],"corresponding":[88],"technique,":[89],"termed":[90],"coarse-grain":[91],"sparsification":[92],"(CGS),":[93],"introduces":[94],"hardware-aware":[95],"sparsity":[96],"during":[97,112],"DNN":[99,123],"training,":[100],"which":[101],"leads":[102],"efficient":[104],"compression":[107],"significant":[109],"reduction":[111],"classification":[113],"without":[114],"losing":[115],"accuracy.":[116],"apply":[118],"proposed":[120],"approach":[121],"for":[125,232],"keyword":[126,171,215],"detection":[127,172,216],"recognition.":[130],"When":[131],"two":[133],"DNNs":[134],"trained":[136],"75%":[138],"weights":[141],"dropped":[142],"classified":[144],"5–6":[146],"bit":[147],"precision,":[149,165],"requirement":[153],"is":[154],"reduced":[155],"95%":[157],"compared":[158],"their":[160],"counterparts":[162],"double":[164],"while":[166],"maintaining":[167],"similar":[168],"performance":[169],"accuracy,":[173],"word":[174],"error":[175,179],"rate,":[176],"sentence":[178],"rate.":[180],"To":[181],"validate":[182],"technique":[184,229],"real":[186],"hardware,":[187],"time-multiplexed":[189],"architecture":[190],"using":[191],"shared":[193],"multiply":[194],"accumulate":[196],"(MAC)":[197],"engine":[198],"was":[199],"implemented":[200],"65nm":[202],"40nm":[204,210],"(LP)":[207],"CMOS.":[208],"at":[211],"0.6":[212],"V,":[213],"consumes":[218,225],"36µW":[219],"recognition":[223],"552µW,":[226],"making":[227],"highly":[230],"suitable":[231],"mobile":[233],"wearable":[235],"devices.":[236]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":1},{"year":2020,"cited_by_count":4},{"year":2019,"cited_by_count":12},{"year":2018,"cited_by_count":7},{"year":2017,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}