Matches in SemOpenAlex for { <https://semopenalex.org/work/W2765671566> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W2765671566 abstract "Trans-dimensional random field language models (TRF LMs) where sentences are modeled as a collection of random fields, have shown close performance with LSTM LMs in speech recognition and are computationally more efficient in inference. However, the training efficiency of neural TRF LMs is not satisfactory, which limits the scalability of TRF LMs on large training corpus. In this paper, several techniques on both model formulation and parameter estimation are proposed to improve the training efficiency and the performance of neural TRF LMs. First, TRFs are reformulated in the form of exponential tilting of a reference distribution. Second, noise-contrastive estimation (NCE) is introduced to jointly estimate the model parameters and normalization constants. Third, we extend the neural TRF LMs by marrying the deep convolutional neural network (CNN) and the bidirectional LSTM into the potential function to extract the deep hierarchical features and bidirectionally sequential features. Utilizing all the above techniques enables the successful and efficient training of neural TRF LMs on a 40x larger training set with only 1/3 training time and further reduces the WER with relative reduction of 4.7% on top of a strong LSTM LM baseline." @default.
- W2765671566 created "2017-11-10" @default.
- W2765671566 creator A5010173604 @default.
- W2765671566 creator A5015829772 @default.
- W2765671566 date "2017-10-29" @default.
- W2765671566 modified "2023-09-24" @default.
- W2765671566 title "Learning neural trans-dimensional random field language models with noise-contrastive estimation" @default.
- W2765671566 cites W1520465330 @default.
- W2765671566 cites W1522301498 @default.
- W2765671566 cites W1591801644 @default.
- W2765671566 cites W1936878994 @default.
- W2765671566 cites W1938755728 @default.
- W2765671566 cites W2158195707 @default.
- W2765671566 cites W2167717037 @default.
- W2765671566 cites W2259472270 @default.
- W2765671566 cites W2289760663 @default.
- W2765671566 cites W2302747106 @default.
- W2765671566 cites W2608207374 @default.
- W2765671566 cites W2747886799 @default.
- W2765671566 cites W2949457404 @default.
- W2765671566 doi "https://doi.org/10.48550/arxiv.1710.10739" @default.
- W2765671566 hasPublicationYear "2017" @default.
- W2765671566 type Work @default.
- W2765671566 sameAs 2765671566 @default.
- W2765671566 citedByCount "0" @default.
- W2765671566 crossrefType "posted-content" @default.
- W2765671566 hasAuthorship W2765671566A5010173604 @default.
- W2765671566 hasAuthorship W2765671566A5015829772 @default.
- W2765671566 hasBestOaLocation W27656715661 @default.
- W2765671566 hasConcept C102248274 @default.
- W2765671566 hasConcept C11413529 @default.
- W2765671566 hasConcept C115961682 @default.
- W2765671566 hasConcept C136886441 @default.
- W2765671566 hasConcept C144024400 @default.
- W2765671566 hasConcept C153180895 @default.
- W2765671566 hasConcept C154945302 @default.
- W2765671566 hasConcept C19165224 @default.
- W2765671566 hasConcept C2776214188 @default.
- W2765671566 hasConcept C28490314 @default.
- W2765671566 hasConcept C32617633 @default.
- W2765671566 hasConcept C41008148 @default.
- W2765671566 hasConcept C50644808 @default.
- W2765671566 hasConcept C81363708 @default.
- W2765671566 hasConcept C99498987 @default.
- W2765671566 hasConceptScore W2765671566C102248274 @default.
- W2765671566 hasConceptScore W2765671566C11413529 @default.
- W2765671566 hasConceptScore W2765671566C115961682 @default.
- W2765671566 hasConceptScore W2765671566C136886441 @default.
- W2765671566 hasConceptScore W2765671566C144024400 @default.
- W2765671566 hasConceptScore W2765671566C153180895 @default.
- W2765671566 hasConceptScore W2765671566C154945302 @default.
- W2765671566 hasConceptScore W2765671566C19165224 @default.
- W2765671566 hasConceptScore W2765671566C2776214188 @default.
- W2765671566 hasConceptScore W2765671566C28490314 @default.
- W2765671566 hasConceptScore W2765671566C32617633 @default.
- W2765671566 hasConceptScore W2765671566C41008148 @default.
- W2765671566 hasConceptScore W2765671566C50644808 @default.
- W2765671566 hasConceptScore W2765671566C81363708 @default.
- W2765671566 hasConceptScore W2765671566C99498987 @default.
- W2765671566 hasLocation W27656715661 @default.
- W2765671566 hasOpenAccess W2765671566 @default.
- W2765671566 hasPrimaryLocation W27656715661 @default.
- W2765671566 hasRelatedWork W1991269640 @default.
- W2765671566 hasRelatedWork W2016839265 @default.
- W2765671566 hasRelatedWork W2018445155 @default.
- W2765671566 hasRelatedWork W2175746458 @default.
- W2765671566 hasRelatedWork W2732542196 @default.
- W2765671566 hasRelatedWork W2760085659 @default.
- W2765671566 hasRelatedWork W2921836287 @default.
- W2765671566 hasRelatedWork W3081496756 @default.
- W2765671566 hasRelatedWork W3093612317 @default.
- W2765671566 hasRelatedWork W3212798775 @default.
- W2765671566 isParatext "false" @default.
- W2765671566 isRetracted "false" @default.
- W2765671566 magId "2765671566" @default.
- W2765671566 workType "article" @default.