Abstract
The new genetic algorithm for training layered feedforward neural networks proposed here uses a mutation operator for performing the search behaviors of local optimization. Combining the random restart method with the local search technique, the algorithm can converge asymptocally, to the optimal solution. Test with a practical example showed that the improved genetic algorthm is more efficient than the conventional genetic algorithm.
References
Buford, D. S., Rackesh, S. 1986. Thermodynamic Data for Pure Compounds. Elsevier, Amsterdam.
Chen, G. L., Wang, X. F. et al, 1996. Genetic algorithm and its application. People’s Post Press, Beijing (in Chinese).
Kitano, H., 1990. Empirical studies on the speed of convergence of neural network training using genetic algorithms,In: Proceedings AAAI-90, p. 789–795.
Montana, D. J., Davis, L., 1989. Training feedforward neutral networks using genetic algorithms.In: Proceedings IJCAI-89, p. 762–767.
Vargaftik, N. B., 1975. Tables on the Thermophysical Properties of Liquids and Gases. Hemisphere Publishing Corporation, Wiley, New York, p. 99–175.
Xu, Z. B., Gao, Y., 1996. The characteristic analysis and prevention of premature convergence of genetic algorithms.Science in China, Ser. E,26(4): 364–375.
Byungjoo, Yoon, Holmes, Dawn J., et al., 1994: Efficient genetic algorithms or training layered feedforward neural networks.Information Sciences,76: 67–85.
Author information
Authors and Affiliations
Additional information
Project supported by NSFC (No 39870940) and (G199054405-973) the National Key Scientific Research & Development Program.
Rights and permissions
About this article
Cite this article
Ping, L., Yi-yu, C. An improved genetic algorithm for training layered feedforward neural networks. J. Zhejiang Univ. Sci. A 1, 322–326 (2000). https://doi.org/10.1631/BF02910644
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1631/BF02910644