Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963640903> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W2963640903 abstract "Trans-dimensional random field language models (TRF LMs) have recently been introduced, where sentences are modeled as a collection of random fields. The TRF approach has been shown to have the advantages of being computationally more efficient in inference than LSTM LMs with close performance and being able to flexibly integrating rich features. In this paper we propose neural TRFs, beyond of the previous discrete TRFs that only use linear potentials with discrete features. The idea is to use nonlinear potentials with continuous features, implemented by neural networks (NNs), in the TRF framework. Neural TRFs combine the advantages of both NNs and TRFs. The benefits of word embedding, nonlinear feature learning and larger context modeling are inherited from the use of NNs. At the same time, the strength of efficient inference by avoiding expensive softmax is preserved. A number of technical contributions, including employing deep convolutional neural networks (CNNs) to define the potentials and incorporating the joint stochastic approximation (JSA) strategy in the training algorithm, are developed in this work, which enable us to successfully train neural TRF LMs. Various LMs are evaluated in terms of speech recognition WERs by rescoring the 1000-best lists of WSJ'92 test data. The results show that neural TRF LMs not only improve over discrete TRF LMs, but also perform slightly better than LSTM LMs with only one fifth of parameters and 16x faster inference efficiency." @default.
- W2963640903 created "2019-07-30" @default.
- W2963640903 creator A5010173604 @default.
- W2963640903 creator A5033732336 @default.
- W2963640903 date "2017-12-01" @default.
- W2963640903 modified "2023-10-16" @default.
- W2963640903 title "Language modeling with neural trans-dimensional random fields" @default.
- W2963640903 cites W1568229137 @default.
- W2963640903 cites W1970689298 @default.
- W2963640903 cites W1981678206 @default.
- W2963640903 cites W1984635093 @default.
- W2963640903 cites W1994616650 @default.
- W2963640903 cites W2076440176 @default.
- W2963640903 cites W2158195707 @default.
- W2963640903 cites W2171928131 @default.
- W2963640903 cites W2251519907 @default.
- W2963640903 cites W2402268235 @default.
- W2963640903 cites W2512007918 @default.
- W2963640903 cites W2566563465 @default.
- W2963640903 cites W2608339501 @default.
- W2963640903 cites W2951559648 @default.
- W2963640903 cites W2962902328 @default.
- W2963640903 doi "https://doi.org/10.1109/asru.2017.8268949" @default.
- W2963640903 hasPublicationYear "2017" @default.
- W2963640903 type Work @default.
- W2963640903 sameAs 2963640903 @default.
- W2963640903 citedByCount "10" @default.
- W2963640903 countsByYear W29636409032018 @default.
- W2963640903 countsByYear W29636409032020 @default.
- W2963640903 countsByYear W29636409032021 @default.
- W2963640903 crossrefType "proceedings-article" @default.
- W2963640903 hasAuthorship W2963640903A5010173604 @default.
- W2963640903 hasAuthorship W2963640903A5033732336 @default.
- W2963640903 hasBestOaLocation W29636409032 @default.
- W2963640903 hasConcept C11413529 @default.
- W2963640903 hasConcept C119857082 @default.
- W2963640903 hasConcept C121332964 @default.
- W2963640903 hasConcept C138885662 @default.
- W2963640903 hasConcept C151730666 @default.
- W2963640903 hasConcept C153180895 @default.
- W2963640903 hasConcept C154945302 @default.
- W2963640903 hasConcept C158622935 @default.
- W2963640903 hasConcept C188441871 @default.
- W2963640903 hasConcept C202444582 @default.
- W2963640903 hasConcept C2776214188 @default.
- W2963640903 hasConcept C2776401178 @default.
- W2963640903 hasConcept C2779343474 @default.
- W2963640903 hasConcept C33923547 @default.
- W2963640903 hasConcept C41008148 @default.
- W2963640903 hasConcept C41608201 @default.
- W2963640903 hasConcept C41895202 @default.
- W2963640903 hasConcept C50644808 @default.
- W2963640903 hasConcept C62520636 @default.
- W2963640903 hasConcept C81363708 @default.
- W2963640903 hasConcept C86803240 @default.
- W2963640903 hasConcept C9652623 @default.
- W2963640903 hasConceptScore W2963640903C11413529 @default.
- W2963640903 hasConceptScore W2963640903C119857082 @default.
- W2963640903 hasConceptScore W2963640903C121332964 @default.
- W2963640903 hasConceptScore W2963640903C138885662 @default.
- W2963640903 hasConceptScore W2963640903C151730666 @default.
- W2963640903 hasConceptScore W2963640903C153180895 @default.
- W2963640903 hasConceptScore W2963640903C154945302 @default.
- W2963640903 hasConceptScore W2963640903C158622935 @default.
- W2963640903 hasConceptScore W2963640903C188441871 @default.
- W2963640903 hasConceptScore W2963640903C202444582 @default.
- W2963640903 hasConceptScore W2963640903C2776214188 @default.
- W2963640903 hasConceptScore W2963640903C2776401178 @default.
- W2963640903 hasConceptScore W2963640903C2779343474 @default.
- W2963640903 hasConceptScore W2963640903C33923547 @default.
- W2963640903 hasConceptScore W2963640903C41008148 @default.
- W2963640903 hasConceptScore W2963640903C41608201 @default.
- W2963640903 hasConceptScore W2963640903C41895202 @default.
- W2963640903 hasConceptScore W2963640903C50644808 @default.
- W2963640903 hasConceptScore W2963640903C62520636 @default.
- W2963640903 hasConceptScore W2963640903C81363708 @default.
- W2963640903 hasConceptScore W2963640903C86803240 @default.
- W2963640903 hasConceptScore W2963640903C9652623 @default.
- W2963640903 hasLocation W29636409031 @default.
- W2963640903 hasLocation W29636409032 @default.
- W2963640903 hasOpenAccess W2963640903 @default.
- W2963640903 hasPrimaryLocation W29636409031 @default.
- W2963640903 hasRelatedWork W2610906757 @default.
- W2963640903 hasRelatedWork W2743258233 @default.
- W2963640903 hasRelatedWork W2758063741 @default.
- W2963640903 hasRelatedWork W2760085659 @default.
- W2963640903 hasRelatedWork W2883041339 @default.
- W2963640903 hasRelatedWork W2883935556 @default.
- W2963640903 hasRelatedWork W2909857627 @default.
- W2963640903 hasRelatedWork W2944095174 @default.
- W2963640903 hasRelatedWork W2977314777 @default.
- W2963640903 hasRelatedWork W3208883981 @default.
- W2963640903 isParatext "false" @default.
- W2963640903 isRetracted "false" @default.
- W2963640903 magId "2963640903" @default.
- W2963640903 workType "article" @default.