BAYESIAN RECURRENT NEURAL NETWORK LANGUAGE MODEL
顯示簡易詮釋資料 統計資料| 欄位 | 內容 | 語言 |
|---|---|---|
dc.citation.epage |
211 | en_US |
dc.citation.spage |
206 | en_US |
dc.citation.woscount |
0 | en_US |
dc.contributor.author |
Chien, Jen-Tzung | en_US |
dc.contributor.author |
Ku, Yuan-Chu | en_US |
dc.contributor.department |
電機學院 | zh_TW |
dc.contributor.department |
College of Electrical and Computer Engineering | en_US |
dc.date.accessioned |
2017-04-21T06:48:53Z | |
dc.date.available |
2017-04-21T06:48:53Z | |
dc.date.issued |
2014 | en_US |
dc.description.abstract |
This paper presents a Bayesian approach to construct the recurrent neural network language model (RNN-LM) for speech recognition. Our idea is to regularize the RNN-LM by compensating the uncertainty of the estimated model parameters which is represented by a Gaussian prior. The objective function in Bayesian RNN (BRNN) is formed as the regularized cross entropy error function. The regularized model is not only constructed by training the regularized parameters according to the maximum a posteriori criterion but also estimating the Gaussian hyperparameter by maximizing the marginal likelihood. A rapid approximation to Hessian matrix is developed by selecting a small set of salient outer-products and illustrated to be effective for BRNN-LM. BRNN-LM achieves sparser model than RNN-LM. Experiments on different corpora show promising improvement by applying BRNN-LM using different amount of training data. | en_US |
dc.identifier.isbn |
978-1-4799-7129-9 | en_US |
dc.identifier.journal |
2014 IEEE WORKSHOP ON SPOKEN LANGUAGE TECHNOLOGY SLT 2014 | en_US |
dc.identifier.uri |
https://ir.lib.nycu.edu.tw/handle/11536/135882 | |
dc.identifier.wosnumber |
WOS:000380375100035 | en_US |
dc.language.iso |
en_US | en_US |
dc.subject |
Recurrent neural network | en_US |
dc.subject |
language model | en_US |
dc.subject |
Bayesian learning | en_US |
dc.subject |
Hessian matrix | en_US |
dc.title |
BAYESIAN RECURRENT NEURAL NETWORK LANGUAGE MODEL | en_US |
dc.type |
Proceedings Paper | en_US |
檔案
本條目沒有附加檔案。
所屬典藏: 會議論文
引用此條目
Chien, J.-T., & Ku, Y.-C. (2014). BAYESIAN RECURRENT NEURAL NETWORK LANGUAGE MODEL. 2014 IEEE WORKSHOP ON SPOKEN LANGUAGE TECHNOLOGY SLT 2014, 206–211. https://ir.lib.nycu.edu.tw/handle/11536/135882
引用文字依 APA 第 7 版格式自動產生,使用前請再核對。
分享
永久網址:https://ir.lib.nycu.edu.tw/handle/11536/135882