Chaos time-series prediction based on an improved recursive Levenberg-Marquardt algorithm

被引：16
作者：
Shi, Xiancheng [1 ,2 ]
Feng, Yucheng [1 ,2 ]
Zeng, Jinsong [1 ,2 ]
Chen, Kefu [1 ,2 ]
机构：
[1] South China Univ Technol, Sch Light Ind & Engn, Guangzhou 510641, Guangdong, Peoples R China
[2] State Key Lab Pulp & Paper Engn, Guangzhou 510641, Guangdong, Peoples R China
来源：
CHAOS SOLITONS & FRACTALS | 2017年 / 100卷
关键词：
Recursive algorithm; Levenberg-Marquardt; On-line leaming; Neural networks; NEURAL-NETWORK;
D O I：
10.1016/j.chaos.2017.04.032
中图分类号：
O1 [数学];
学科分类号：
0701 ; 070101 ;
摘要：
An improved recursive Levenberg-Marquardt algorithm (RLM) is proposed to more efficiently train neural networks. The error criterion of the RLM algorithm was modified to reduce the impact of the forgetting factor on the convergence of the algorithm. The remedy to apply the matrix inversion lemma in the RLM algorithm was extended from one row to multiple rows to improve the success rate of the convergence; after that, the adjustment strategy was modified based on the extended remedy. Finally, the performance of this algorithm was tested on two chaotic systems. The results show improved convergence. (C) 2017 Elsevier Ltd. All rights reserved.
引用
收藏
页码：57 / 61
页数：5
相关论文
共 22 条
  • [1] Two highly efficient second-order algorithms for training feedforward networks
    Ampazis, N
    Perantonis, SJ
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2002, 13 (05): : 1064 - 1074
  • [2] Asirvadam VS, 2002, NEURAL NETWORKS FOR SIGNAL PROCESSING XII, PROCEEDINGS, P129, DOI 10.1109/NNSP.2002.1030024
  • [3] Improving generalization of MLPs with sliding mode control and the Levenberg-Marquardt algorithm
    Costa, Marcelo Azevedo
    Braga, Antonio de Padua
    de Menezes, Benjamin Rodrigues
    [J]. NEUROCOMPUTING, 2007, 70 (7-9) : 1342 - 1347
  • [4] Daily reservoir inflow forecasting using artificial neural networks with stopped training approach
    Coulibaly, P
    Anctil, F
    Bobée, B
    [J]. JOURNAL OF HYDROLOGY, 2000, 230 (3-4) : 244 - 257
  • [5] A sliding window solution for the on-line implementation of the Levenberg-Marquardt algorithm
    Dias, FM
    Antunes, A
    Vieira, J
    Mota, A
    [J]. ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2006, 19 (01) : 1 - 7
  • [6] Online Sliding-Window Methods for Process Model Adaptation
    Ferreira, Pedro M.
    Ruano, Antonio E.
    [J]. IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2009, 58 (09) : 3012 - 3020
  • [7] Golub GH., 2012, MATRIX COMPUTATIONS, V3
  • [8] Second-order training of adaptive critics for online process control
    Govindhasamy, JJ
    McLoone, SF
    Irwin, GW
    [J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2005, 35 (02): : 381 - 385
  • [9] Fuzzy prediction architecture using recurrent neural networks
    Graves, Daniel
    Pedrycz, Witold
    [J]. NEUROCOMPUTING, 2009, 72 (7-9) : 1668 - 1678
  • [10] TRAINING FEEDFORWARD NETWORKS WITH THE MARQUARDT ALGORITHM
    HAGAN, MT
    MENHAJ, MB
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 1994, 5 (06): : 989 - 993
123