期刊
NEURAL NETWORKS
卷 16, 期 2, 页码 241-250出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/S0893-6080(02)00219-8
关键词
long short-term memory; recurrent neural networks; decoupled extended Kalman filter; online prediction; context sensitive language inference
The long short-term memory (LSTM) network trained by gradient descent solves difficult problems which traditional recurrent neural networks in general cannot. We have recently observed that the decoupled extended Kalman filter training algorithm allows for even better performance, reducing significantly the number of training steps when compared to the original gradient descent training algorithm. In this paper we present a set of experiments which are unsolvable by classical recurrent networks but which are solved elegantly and robustly and quickly by LSTM combined with Kalman filters. (C) 2003 Elsevier Science Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据