An improved recursive prediction error algorithm for training recurrent neural networks

被引：0
|
作者：
Li, HR [1 ]
Wang, XZ [1 ]
Gu, SS [1 ]
机构：
[1] NE Univ Technol, Sch Informat Sci & Engn, Shenyang 110006, Peoples R China
来源：
PROCEEDINGS OF THE 3RD WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION, VOLS 1-5 | 2000年
关键词：
D O I：
暂无
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
In this paper, a fast and effective learning algorithm for training recurrent neural networks, which is realized by introducing and improving the recursive prediction error (RPE) method, is proposed. The improving scheme for RPE algorithm is adding momentum term in the gradient of Gauss-Newton search direction and using changeable forgetting factor Simulation results show that the proposed algorithm achieves far better convergence performance than the classical hack propagation with momentum term algorithm, and has superior performance compared with the conventional RPE algorithm.
引用
收藏
页码：1043 / 1046
页数：4
相关论文
共 50 条
  • [1] Training recurrent neural networks by using parallel recursive prediction error algorithm
    Chen, DQ
    Chan, LW
    [J]. ICONIP'98: THE FIFTH INTERNATIONAL CONFERENCE ON NEURAL INFORMATION PROCESSING JOINTLY WITH JNNS'98: THE 1998 ANNUAL CONFERENCE OF THE JAPANESE NEURAL NETWORK SOCIETY - PROCEEDINGS, VOLS 1-3, 1998, : 1393 - 1396
  • [2] PARALLEL RECURSIVE PREDICTION ERROR ALGORITHM FOR TRAINING LAYERED NEURAL NETWORKS
    CHEN, S
    COWAN, CFN
    BILLINGS, SA
    GRANT, PM
    [J]. INTERNATIONAL JOURNAL OF CONTROL, 1990, 51 (06) : 1215 - 1228
  • [3] A second order recursive prediction error algorithm for diagonal recurrent neural networks
    Wang, YJ
    Fernholz, G
    Engell, S
    [J]. PROCEEDINGS OF THE 1998 IEEE INTERNATIONAL CONFERENCE ON CONTROL APPLICATIONS, VOLS 1 AND 2, 1996, : 172 - 176
  • [4] Training of neural network Wiener models with recursive prediction error algorithm
    Janczak, A
    [J]. NEURAL NETWORKS AND SOFT COMPUTING, 2003, : 692 - 697
  • [5] A COMPARISON OF THE BACKPROPAGATION AND RECURSIVE PREDICTION ERROR ALGORITHMS FOR TRAINING NEURAL NETWORKS
    BILLINGS, SA
    JAMALUDDIN, HB
    CHEN, S
    [J]. MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 1991, 5 (03) : 233 - 255
  • [6] A fast and robust recursive prediction error learning algorithm for feedforward neural networks
    Zhang, YM
    Li, XR
    [J]. PROCEEDINGS OF THE 35TH IEEE CONFERENCE ON DECISION AND CONTROL, VOLS 1-4, 1996, : 2036 - 2041
  • [7] A training algorithm for partial recurrent neural networks
    Al-Faysale, MSM
    [J]. Modelling and Simulation 2004, 2004, : 350 - 354
  • [8] A learning algorithm for improved recurrent neural networks
    Chen, CH
    Yu, LW
    [J]. 1997 IEEE INTERNATIONAL CONFERENCE ON NEURAL NETWORKS, VOLS 1-4, 1997, : 2198 - 2202
  • [9] IMPROVED RESULTS FOR COLOR ERROR REDUCTIONS USING NEURAL NETWORKS AND A FAST TRAINING ALGORITHM
    CRILLY, PB
    XU, SF
    [J]. IEEE TRANSACTIONS ON CONSUMER ELECTRONICS, 1993, 39 (04) : 923 - 927
  • [10] Recursive Bayesian Levenberg-Marquardt training of recurrent neural networks
    Mirikitani, Derrick
    Nikolaev, Nikolay
    [J]. 2007 IEEE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOLS 1-6, 2007, : 282 - 287
12345