{"id":"https://openalex.org/W2896144487","doi":"https://doi.org/10.1109/ijcnn.2018.8489156","title":"Fast Training and Model Compression of Gated RNNs via Singular Value Decomposition","display_name":"Fast Training and Model Compression of Gated RNNs via Singular Value Decomposition","publication_year":2018,"publication_date":"2018-07-01","ids":{"openalex":"https://openalex.org/W2896144487","doi":"https://doi.org/10.1109/ijcnn.2018.8489156","mag":"2896144487"},"language":"en","primary_location":{"id":"doi:10.1109/ijcnn.2018.8489156","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn.2018.8489156","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5051590067","display_name":"Rui Dai","orcid":"https://orcid.org/0000-0002-3698-4086"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Rui Dai","raw_affiliation_strings":["Dept. Industrial Engineering, Tsinghua University, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dept. Industrial Engineering, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021941275","display_name":"Lefei Li","orcid":"https://orcid.org/0000-0001-6816-6267"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lefei Li","raw_affiliation_strings":["Dept. Industrial Engineering, Tsinghua University, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dept. Industrial Engineering, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5053437305","display_name":"Wenjian Yu","orcid":"https://orcid.org/0000-0003-4897-7251"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenjian Yu","raw_affiliation_strings":["BNRist, Dept. Computer Science & Tech., Tsinghua University, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"BNRist, Dept. Computer Science & Tech., Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":0.6845,"has_fulltext":false,"cited_by_count":12,"citation_normalized_percentile":{"value":0.82328175,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9993000030517578,"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":0.9993000030517578,"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/T10320","display_name":"Neural Networks and Applications","score":0.9991999864578247,"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/T12676","display_name":"Machine Learning and ELM","score":0.9984999895095825,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/singular-value-decomposition","display_name":"Singular value decomposition","score":0.8621822595596313},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.8177019357681274},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.7849705815315247},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7516034841537476},{"id":"https://openalex.org/keywords/mnist-database","display_name":"MNIST database","score":0.7279723882675171},{"id":"https://openalex.org/keywords/matrix-decomposition","display_name":"Matrix decomposition","score":0.6202359795570374},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5218176245689392},{"id":"https://openalex.org/keywords/rank","display_name":"Rank (graph theory)","score":0.5080623626708984},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.44496962428092957},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.43286603689193726},{"id":"https://openalex.org/keywords/factorization","display_name":"Factorization","score":0.4253199100494385},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3512178659439087},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1441725790500641},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.12342873215675354}],"concepts":[{"id":"https://openalex.org/C22789450","wikidata":"https://www.wikidata.org/wiki/Q420904","display_name":"Singular value decomposition","level":2,"score":0.8621822595596313},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.8177019357681274},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.7849705815315247},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7516034841537476},{"id":"https://openalex.org/C190502265","wikidata":"https://www.wikidata.org/wiki/Q17069496","display_name":"MNIST database","level":3,"score":0.7279723882675171},{"id":"https://openalex.org/C42355184","wikidata":"https://www.wikidata.org/wiki/Q1361088","display_name":"Matrix decomposition","level":3,"score":0.6202359795570374},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5218176245689392},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.5080623626708984},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.44496962428092957},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.43286603689193726},{"id":"https://openalex.org/C187834632","wikidata":"https://www.wikidata.org/wiki/Q188804","display_name":"Factorization","level":2,"score":0.4253199100494385},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3512178659439087},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1441725790500641},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.12342873215675354},{"id":"https://openalex.org/C158693339","wikidata":"https://www.wikidata.org/wiki/Q190524","display_name":"Eigenvalues and eigenvectors","level":2,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ijcnn.2018.8489156","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn.2018.8489156","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":49,"referenced_works":["https://openalex.org/W1486541123","https://openalex.org/W1522301498","https://openalex.org/W1536680647","https://openalex.org/W1593564458","https://openalex.org/W1924770834","https://openalex.org/W1985802984","https://openalex.org/W2004026774","https://openalex.org/W2064675550","https://openalex.org/W2095705004","https://openalex.org/W2098502158","https://openalex.org/W2112796928","https://openalex.org/W2113459411","https://openalex.org/W2119144962","https://openalex.org/W2133564696","https://openalex.org/W2143017621","https://openalex.org/W2143612262","https://openalex.org/W2161591461","https://openalex.org/W2168231600","https://openalex.org/W2182629226","https://openalex.org/W2198190323","https://openalex.org/W2271840356","https://openalex.org/W2279057335","https://openalex.org/W2281572220","https://openalex.org/W2402268235","https://openalex.org/W2605203995","https://openalex.org/W2949964376","https://openalex.org/W2952899695","https://openalex.org/W2963674932","https://openalex.org/W2964121744","https://openalex.org/W2964200805","https://openalex.org/W2964299589","https://openalex.org/W2964308564","https://openalex.org/W4251372957","https://openalex.org/W4298074540","https://openalex.org/W6631190155","https://openalex.org/W6635440621","https://openalex.org/W6640212811","https://openalex.org/W6674330103","https://openalex.org/W6676984168","https://openalex.org/W6677580257","https://openalex.org/W6679434410","https://openalex.org/W6683826617","https://openalex.org/W6684859321","https://openalex.org/W6686067075","https://openalex.org/W6687785706","https://openalex.org/W6694517276","https://openalex.org/W6695321055","https://openalex.org/W6713098461","https://openalex.org/W6736198902"],"related_works":["https://openalex.org/W2782904003","https://openalex.org/W2118633810","https://openalex.org/W2150953077","https://openalex.org/W4226434912","https://openalex.org/W2002598339","https://openalex.org/W1995410415","https://openalex.org/W2133814741","https://openalex.org/W3108343813","https://openalex.org/W4312481107","https://openalex.org/W2482295163"],"abstract_inverted_index":{"Long":[0],"Short-Term":[1],"Memory":[2],"(LSTM)":[3],"network":[4,10,140],"and":[5,29,59,81,86,100,107],"Gated":[6],"Recurrent":[7,15],"Units":[8],"(GRU)":[9],"are":[11,64,93],"two":[12],"widely-used":[13],"gated":[14,90],"Neural":[16],"Network":[17],"(RNN)":[18],"architectures.":[19],"Both":[20],"of":[21,53,88,132,153],"them":[22],"usually":[23],"have":[24,142],"a":[25,31,43],"huge":[26],"model":[27,58,80],"size":[28],"require":[30],"long":[32],"time":[33,85],"to":[34,120],"be":[35,158],"trained.":[36],"In":[37],"this":[38],"paper,":[39],"we":[40],"first":[41],"propose":[42],"singular":[44],"value":[45],"decomposition":[46],"(SVD)":[47],"based":[48,61],"approach":[49,63,117],"for":[50,66,76,136],"fast":[51],"training":[52,62,84,125,137],"LSTM.":[54],"Then,":[55],"the":[56,67,73,77,83,89,97,114,126,138,148,151],"factorized":[57,149],"SVD":[60],"proposed":[65,115],"GRU":[68,139],"network,":[69],"which":[70],"adaptively":[71],"choose":[72],"rank":[74],"parameter":[75],"matrix":[78],"factorization":[79],"reduce":[82],"parameters":[87,156],"RNNs.":[91],"Experiments":[92],"carried":[94],"out":[95],"on":[96],"image":[98],"classification":[99,102],"sentiment":[101],"tasks":[103],"using":[104],"datasets":[105],"MNIST":[106],"IMDB,":[108],"respectively.":[109],"The":[110,134],"results":[111],"show":[112],"that":[113],"LSTM-SVD":[116],"achieves":[118],"up":[119],"3.9X":[121],"speedup":[122],"compared":[123],"with":[124,147],"original":[127],"LSTM":[128],"model,":[129],"without":[130],"loss":[131],"accuracy.":[133],"approaches":[135],"also":[141],"about":[143],"2X":[144],"speedup.":[145],"And,":[146],"models":[150],"quantity":[152],"RNN":[154],"cell":[155],"can":[157],"significantly":[159],"reduced":[160],"by":[161],"more":[162],"than":[163],"10X.":[164]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":5},{"year":2020,"cited_by_count":2},{"year":2019,"cited_by_count":1},{"year":2018,"cited_by_count":1}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}