{"id":"https://openalex.org/W2962780420","doi":"https://doi.org/10.21437/interspeech.2018-2335","title":"Combining Natural Gradient with Hessian Free Methods for Sequence Training","display_name":"Combining Natural Gradient with Hessian Free Methods for Sequence Training","publication_year":2018,"publication_date":"2018-08-28","ids":{"openalex":"https://openalex.org/W2962780420","doi":"https://doi.org/10.21437/interspeech.2018-2335","mag":"2962780420"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2018-2335","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2018-2335","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2018","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/A5006496564","display_name":"Adnan Haider","orcid":"https://orcid.org/0000-0002-7075-1796"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]},{"id":"https://openalex.org/I4210096386","display_name":"Bridge University","ror":"https://ror.org/00cbm0437","country_code":"SS","type":"education","lineage":["https://openalex.org/I4210096386"]}],"countries":["GB","SS"],"is_corresponding":false,"raw_author_name":"Adnan Haider","raw_affiliation_strings":["Cambridge University Engineering Dept., Trumpington St., Cambridge, CB2 1PZ U.K"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Cambridge University Engineering Dept., Trumpington St., Cambridge, CB2 1PZ U.K","institution_ids":["https://openalex.org/I241749","https://openalex.org/I4210096386"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5002191410","display_name":"Philip C. Woodland","orcid":"https://orcid.org/0000-0001-9069-0225"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]},{"id":"https://openalex.org/I4210096386","display_name":"Bridge University","ror":"https://ror.org/00cbm0437","country_code":"SS","type":"education","lineage":["https://openalex.org/I4210096386"]}],"countries":["GB","SS"],"is_corresponding":false,"raw_author_name":"Philip Woodland","raw_affiliation_strings":["Cambridge University Engineering Dept., Trumpington St., Cambridge, CB2 1PZ U.K"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Cambridge University Engineering Dept., Trumpington St., Cambridge, CB2 1PZ U.K","institution_ids":["https://openalex.org/I241749","https://openalex.org/I4210096386"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2918","last_page":"2922"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"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":0.9998999834060669,"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":0.9998999834060669,"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":0.9997000098228455,"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/sigmoid-function","display_name":"Sigmoid function","score":0.7183963060379028},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7143628597259521},{"id":"https://openalex.org/keywords/hessian-matrix","display_name":"Hessian matrix","score":0.7038097381591797},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6176599264144897},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.5977293848991394},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.555852472782135},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.5534595847129822},{"id":"https://openalex.org/keywords/discriminative-model","display_name":"Discriminative model","score":0.5371710658073425},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.5311775207519531},{"id":"https://openalex.org/keywords/pooling","display_name":"Pooling","score":0.49914050102233887},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.4583098888397217},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4090811014175415},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.37602663040161133},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.3432157635688782},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.342135488986969},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.20514532923698425},{"id":"https://openalex.org/keywords/applied-mathematics","display_name":"Applied mathematics","score":0.11530786752700806}],"concepts":[{"id":"https://openalex.org/C81388566","wikidata":"https://www.wikidata.org/wiki/Q526668","display_name":"Sigmoid function","level":3,"score":0.7183963060379028},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7143628597259521},{"id":"https://openalex.org/C203616005","wikidata":"https://www.wikidata.org/wiki/Q620495","display_name":"Hessian matrix","level":2,"score":0.7038097381591797},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6176599264144897},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.5977293848991394},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.555852472782135},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.5534595847129822},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.5371710658073425},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.5311775207519531},{"id":"https://openalex.org/C70437156","wikidata":"https://www.wikidata.org/wiki/Q7228652","display_name":"Pooling","level":2,"score":0.49914050102233887},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.4583098888397217},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4090811014175415},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.37602663040161133},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3432157635688782},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.342135488986969},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.20514532923698425},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.11530786752700806},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C54355233","wikidata":"https://www.wikidata.org/wiki/Q7162","display_name":"Genetics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21437/interspeech.2018-2335","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2018-2335","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2018","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","score":0.7599999904632568,"display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W38527073","https://openalex.org/W1665214252","https://openalex.org/W1844261860","https://openalex.org/W2003123121","https://openalex.org/W2071310251","https://openalex.org/W2095705004","https://openalex.org/W2098841537","https://openalex.org/W2130984546","https://openalex.org/W2131342762","https://openalex.org/W2150907703","https://openalex.org/W2158069733","https://openalex.org/W2160815625","https://openalex.org/W2167729035","https://openalex.org/W2288371874","https://openalex.org/W2288402171","https://openalex.org/W2402146185","https://openalex.org/W2407698769","https://openalex.org/W2436219157","https://openalex.org/W2485135680","https://openalex.org/W2514741789","https://openalex.org/W2794924060","https://openalex.org/W2962765220","https://openalex.org/W2963855604","https://openalex.org/W2963857170","https://openalex.org/W4251193772","https://openalex.org/W4251416849","https://openalex.org/W4297797495"],"related_works":["https://openalex.org/W4385957115","https://openalex.org/W2611031068","https://openalex.org/W1704347466","https://openalex.org/W4283017538","https://openalex.org/W2061372042","https://openalex.org/W3143650729","https://openalex.org/W3177326532","https://openalex.org/W2042173174","https://openalex.org/W4287755480","https://openalex.org/W2785875001"],"abstract_inverted_index":{"This":[0],"paper":[1],"presents":[2],"a":[3,53,86,134],"new":[4],"optimisation":[5,58],"approach":[6,115],"to":[7,45,60,133],"train":[8,62],"Deep":[9],"Neural":[10,70],"Networks":[11,71],"(DNNs)":[12],"with":[13,29],"discriminative":[14],"sequence":[15,61],"criteria.At":[16],"each":[17],"iteration,":[18],"the":[19,24,34,42,79,108,120],"method":[20,48,80],"combines":[21],"information":[22,32],"from":[23],"Natural":[25],"Gradient":[26],"(NG)":[27],"direction":[28],"local":[30],"curvature":[31],"of":[33,78,111],"error":[35,122],"surface":[36],"that":[37,106],"enables":[38],"better":[39],"paths":[40],"on":[41,85],"parameter":[43],"manifold":[44],"be":[46],"traversed.The":[47],"has":[49],"been":[50,102],"applied":[51],"within":[52],"Hessian":[54],"Free":[55],"(HF)":[56],"style":[57],"framework":[59],"both":[63,126],"standard":[64,138],"fully-connected":[65],"DNNs":[66],"and":[67,92,97,128,130],"Time":[68],"Delay":[69],"as":[72],"speech":[73],"recognition":[74],"acoustic":[75],"models.The":[76],"efficacy":[77],"is":[81,104],"shown":[82,105],"using":[83,95],"experiments":[84],"Multi-Genre":[87],"Broadcast":[88],"(MGB)":[89],"transcription":[90],"task":[91],"neural":[93],"networks":[94],"sigmoid":[96],"ReLU":[98],"activation":[99],"functions":[100],"have":[101],"investigated.It":[103],"for":[107],"same":[109],"number":[110],"updates":[112],"this":[113],"proposed":[114],"achieves":[116],"larger":[117],"reductions":[118],"in":[119],"word":[121],"rate":[123],"(WER)":[124],"than":[125,137],"NG":[127],"HF,":[129],"also":[131],"leads":[132],"lower":[135],"WER":[136],"stochastic":[139],"gradient":[140],"descent.":[141]},"counts_by_year":[{"year":2021,"cited_by_count":1},{"year":2019,"cited_by_count":1}],"updated_date":"2026-07-22T07:51:19.307946","created_date":"2025-10-10T00:00:00"}