FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W619799124> ?p ?o ?g. }

• W619799124 abstract "Multilingual deep neural networks (DNNs) can act as deep feature extractors and have been applied successfully to crosslanguage acoustic modeling. Learning these feature extractors becomes an expensive task, because of the enlarged multilingual training data and the sequential nature of stochastic gradient descent (SGD). This paper investigates strategies to accelerate the learning process over multiple GPU cards. We propose the DistModel and DistLang frameworks which distribute feature extractor learning by models and languages respectively. The time-synchronous DistModel has the nice property of tolerating infrequent model averaging. With 3 GPUs, DistModel achieves 2.6× speed-up and causes no loss on word error rates. When using DistLang, we observe better acceleration but worse recognition performance. Further evaluations are conducted to scale DistModel to more languages and GPU cards." @default.
• W619799124 created "2016-06-24" @default.
• W619799124 creator A5003251887 @default.
• W619799124 creator A5049341927 @default.
• W619799124 creator A5085262529 @default.
• W619799124 date "2014-09-14" @default.
• W619799124 modified "2023-09-24" @default.
• W619799124 title "Distributed learning of multilingual DNN feature extractors using GPUs" @default.
• W619799124 cites W1218987319 @default.
• W619799124 cites W1661756259 @default.
• W619799124 cites W17156489 @default.
• W619799124 cites W1929850838 @default.
• W619799124 cites W1970088388 @default.
• W619799124 cites W1978660892 @default.
• W619799124 cites W2002342963 @default.
• W619799124 cites W2023155336 @default.
• W619799124 cites W2025198378 @default.
• W619799124 cites W2025487094 @default.
• W619799124 cites W2026369565 @default.
• W619799124 cites W2071310251 @default.
• W619799124 cites W2087402357 @default.
• W619799124 cites W2089917322 @default.
• W619799124 cites W2107469355 @default.
• W619799124 cites W2112739286 @default.
• W619799124 cites W2119531180 @default.
• W619799124 cites W2120209245 @default.
• W619799124 cites W2120480077 @default.
• W619799124 cites W2144839971 @default.
• W619799124 cites W2147768505 @default.
• W619799124 cites W2155273149 @default.
• W619799124 cites W2160306971 @default.
• W619799124 cites W2166706236 @default.
• W619799124 cites W2168231600 @default.
• W619799124 cites W2253807446 @default.
• W619799124 cites W2394932179 @default.
• W619799124 cites W2397534874 @default.
• W619799124 cites W2399153411 @default.
• W619799124 cites W2962719052 @default.
• W619799124 cites W2964138484 @default.
• W619799124 cites W319941341 @default.
• W619799124 cites W567546468 @default.
• W619799124 cites W587565084 @default.
• W619799124 cites W596888916 @default.
• W619799124 cites W655723668 @default.
• W619799124 doi "https://doi.org/10.21437/interspeech.2014-211" @default.
• W619799124 hasPublicationYear "2014" @default.
• W619799124 type Work @default.
• W619799124 sameAs 619799124 @default.
• W619799124 citedByCount "24" @default.
• W619799124 countsByYear W6197991242014 @default.
• W619799124 countsByYear W6197991242015 @default.
• W619799124 countsByYear W6197991242016 @default.
• W619799124 countsByYear W6197991242017 @default.
• W619799124 countsByYear W6197991242018 @default.
• W619799124 countsByYear W6197991242019 @default.
• W619799124 countsByYear W6197991242020 @default.
• W619799124 countsByYear W6197991242021 @default.
• W619799124 countsByYear W6197991242022 @default.
• W619799124 crossrefType "proceedings-article" @default.
• W619799124 hasAuthorship W619799124A5003251887 @default.
• W619799124 hasAuthorship W619799124A5049341927 @default.
• W619799124 hasAuthorship W619799124A5085262529 @default.
• W619799124 hasBestOaLocation W6197991242 @default.
• W619799124 hasConcept C108583219 @default.
• W619799124 hasConcept C111472728 @default.
• W619799124 hasConcept C117978034 @default.
• W619799124 hasConcept C119857082 @default.
• W619799124 hasConcept C127413603 @default.
• W619799124 hasConcept C138885662 @default.
• W619799124 hasConcept C154945302 @default.
• W619799124 hasConcept C162324750 @default.
• W619799124 hasConcept C173608175 @default.
• W619799124 hasConcept C187736073 @default.
• W619799124 hasConcept C189950617 @default.
• W619799124 hasConcept C199360897 @default.
• W619799124 hasConcept C206688291 @default.
• W619799124 hasConcept C21880701 @default.
• W619799124 hasConcept C2776401178 @default.
• W619799124 hasConcept C2780451532 @default.
• W619799124 hasConcept C2984842247 @default.
• W619799124 hasConcept C41008148 @default.
• W619799124 hasConcept C41895202 @default.
• W619799124 hasConcept C50644808 @default.
• W619799124 hasConcept C68339613 @default.
• W619799124 hasConcept C90805587 @default.
• W619799124 hasConcept C98045186 @default.
• W619799124 hasConceptScore W619799124C108583219 @default.
• W619799124 hasConceptScore W619799124C111472728 @default.
• W619799124 hasConceptScore W619799124C117978034 @default.
• W619799124 hasConceptScore W619799124C119857082 @default.
• W619799124 hasConceptScore W619799124C127413603 @default.
• W619799124 hasConceptScore W619799124C138885662 @default.
• W619799124 hasConceptScore W619799124C154945302 @default.
• W619799124 hasConceptScore W619799124C162324750 @default.
• W619799124 hasConceptScore W619799124C173608175 @default.
• W619799124 hasConceptScore W619799124C187736073 @default.
• W619799124 hasConceptScore W619799124C189950617 @default.
• W619799124 hasConceptScore W619799124C199360897 @default.
• W619799124 hasConceptScore W619799124C206688291 @default.
• W619799124 hasConceptScore W619799124C21880701 @default.

• next