Art

J-GLOBAL ID：201902230177259129   Reference number：19A1956554

Gated recurrent unitの局所安定化による勾配爆発の抑制

Author (5)： KANAI SEKITOSHI (NTT ソフトウェアイノベーションセ) ,  KANAI SEKITOSHI (慶応大 理工) ,  FUJIWARA YASUHIRO (NTT ソフトウェアイノベーションセ) ,  IWAMURA SOTETSU (北大 産学・地域協働推進機構) ,  ADACHI SHUICHI (慶応大 理工)
Material： 電子情報通信学会論文誌 D(Web)
Volume： J102-D  Issue： 8  Page： 530-541 (WEB ONLY)  Publication year： Aug. 01, 2019 
JST Material Number： U0473A  ISSN： 1881-0225  Document type： Article
Article type： 原著論文  Country of issue： Japan (JPN)  Language： JAPANESE (JA)

Thesaurus term： *regression model ,  neural network (network) ,  gradient ,  *Machine Learning ,  data ,  Language ,  Music ,  longitudinal data ,  *deep learning ,  recurrent neural network
Semi thesaurus term： *ゲート付き回帰型ユニット ,  recurrent neural network

JST classification (1)： Artificial intelligence  (JE08000Z)

Reference (44)：

• A. Graves, A.-R. Mohamed, and G. Hinton, “Speech recognition with deep recurrent neural networks,” Proc. IEEE International Conf. on Acoustics, Speech, and Signal Processing, pp.6645-6649, May 2013.
• T. Mikolov, Statistical Language Models Based on Neural Networks, PhD thesis, Brno University of Technology, 2012.
• A. Graves and J. Schmidhuber, “Offline handwriting recognition with multidimensional recurrent neural networks,” Proc. Annual Conf. on Neural Information Processing Systems, pp.545-552, Dec. 2009.
• R. Pascanu, T. Mikolov, and Y. Bengio, “On the difficulty of training recurrent neural networks,” Proc. International Conf. on Machine Learning, pp.1310-1318, June 2013.
• Y. Bengio, P. Simard, and P. Frasconi, “Learning long-term dependencies with gradient descent is difficult,” IEEE Trans. Neural Netw., vol.5, no.2, pp.157-166, Jan. 1994.

Terms in the title (3)： 安定化 ,  勾配 ,  爆発