Title: An advanced coevolution scheme for optimisation of artificial neural network

Authors: Dan Zhang; Zhen Gao

Addresses: Faculty of Engineering and Applied Science, University of Ontario Institute of Technology, Oshawa, ON, L1H 7K4, Canada; School of Mechanical Engineering, DongHua University, Shanghai 200051, China ' Faculty of Engineering and Applied Science, University of Ontario Institute of Technology, Oshawa, ON, L1H 7K4, Canada

Abstract: In this research, the optimisation method of artificial neural network based on cooperative coevolution algorithm is proposed aiming at conquering the defects of general genetic algorithm. The division of population, encoding method of neural network, algorithm procedures and algorithm comparison are proposed. The credibility and the validity of the proposed algorithm are verified through various simulations. The results show that the optimisation of neural network based on cooperative coevolution algorithm has the distinct advantage upon the general genetic algorithm.

Keywords: cooperative coevolution; genetic algorithms; artificial neural networks; ANNs; ANN optimisation; simulation.

DOI: 10.1504/IJMRS.2013.057309

International Journal of Mechanisms and Robotic Systems, 2013 Vol.1 No.4, pp.316 - 330

Received: 06 Feb 2013
Accepted: 14 Mar 2013
Published online: 12 Jul 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article