Improved variable-step backpropagation training method for feedforward neural networks

被引：0
|
作者：
Cogswell, R [1 ]
机构：
[1] Monmouth Coll, Dept Math & Comp Sci, Monmouth, IL 61462 USA
来源：
INTERNATIONAL SOCIETY FOR COMPUTERS AND THEIR APPLICATIONS 13TH INTERNATIONAL CONFERENCE ON COMPUTERS AND THEIR APPLICATIONS | 1998年
关键词：
D O I：
暂无
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
Backpropagation is a standard method used for training neural networks. However, such gradient-descent methods suffer from slow convergence, especially as the error tolerance is decreased. A variable step size method is described based on an approximation of the second derivative in the negative gradient direction. The algorithm is shown to reduce the number of iterations and the total run time for high precision learning in small neural networks.
引用
收藏
页码：330 / 335
页数：6
相关论文
共 50 条
  • [1] A Modified Backpropagation Training Algorithm for Feedforward Neural Networks*
    T. Kathirvalavakumar
    P. Thangavel
    Neural Processing Letters, 2006, 23 : 111 - 119
  • [2] A modified backpropagation training algorithm for feedforward neural networks
    Kathirvalavakumar, T
    Thangavel, P
    NEURAL PROCESSING LETTERS, 2006, 23 (02) : 111 - 119
  • [3] Improved Particle Swarm Optimization Combined with Backpropagation for Feedforward Neural Networks
    Han, Fei
    Zhu, Jian-Sheng
    INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS, 2013, 28 (03) : 271 - 288
  • [4] Improved training rules for multilayered feedforward neural networks
    Sung, SW
    Lee, TY
    Park, S
    INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH, 2003, 42 (06) : 1275 - 1278
  • [5] A New Two-Step Gradient-Based Backpropagation Training Method for Neural Networks
    Mu, Xuewen
    Zhang, Yaling
    ADVANCES IN NEURAL NETWORKS - ISNN 2010, PT 1, PROCEEDINGS, 2010, 6063 : 95 - +
  • [6] Step acceleration based training algorithm for feedforward neural networks
    Li, YL
    Wang, KQ
    Zhang, D
    16TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL II, PROCEEDINGS, 2002, : 84 - 87
  • [7] Optimization of a Stirling Engine by Variable-Step Simplified Conjugate-Gradient Method and Neural Network Training Algorithm
    Cheng, Chin-Hsiang
    Lin, Yu-Ting
    ENERGIES, 2020, 13 (19)
  • [8] Exponentiated backpropagation algorithm for multilayer feedforward neural networks
    Srinivasan, N
    Ravichandran, V
    Chan, KL
    Vidhya, JR
    Ramakirishnan, S
    Krishnan, SM
    ICONIP'02: PROCEEDINGS OF THE 9TH INTERNATIONAL CONFERENCE ON NEURAL INFORMATION PROCESSING: COMPUTATIONAL INTELLIGENCE FOR THE E-AGE, 2002, : 327 - 331
  • [9] A general backpropagation algorithm for feedforward neural networks learning
    Yu, XH
    Efe, MO
    Kaynak, O
    IEEE TRANSACTIONS ON NEURAL NETWORKS, 2002, 13 (01): : 251 - 254
  • [10] Feedforward backpropagation artificial neural networks on reconfigurable meshes
    Jenq, JJ
    Li, WN
    FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE, 1998, 14 (5-6): : 313 - 319
12345