Structure optimization strategy of neural networks - research on the pruning algorithm

We analysis the major factors which influence the generalization performance of the neural networks, carry out a survey on structure optimization methods and present several kinds of pruning algorithms to simulate the multilayer feedforward network on two kinds of typical problems. The result sugges...

Full description

Saved in:
Bibliographic Details
Published inProceedings of the 3rd World Congress on Intelligent Control and Automation (Cat. No.00EX393) Vol. 2; pp. 877 - 881 vol.2
Main Authors Zhang Meng, Xu YongMao
Format Conference Proceeding
LanguageEnglish
Published IEEE 2000
Subjects
Algorithm design and analysis
Analytical models
Automation
Neural networks
Optimization methods
Performance analysis
Online AccessGet full text

Cover

More Information
Summary:We analysis the major factors which influence the generalization performance of the neural networks, carry out a survey on structure optimization methods and present several kinds of pruning algorithms to simulate the multilayer feedforward network on two kinds of typical problems. The result suggests that the pruning algorithm can effectively optimize the structure of neural networks and notably improve their generalization. Finally, some new developments of the pruning algorithm are summarized.
ISBN:078035995X
9780780359956
DOI:10.1109/WCICA.2000.863357