Paper

Evolutionary Algorithms for the Complex Network Based on Granular Neural Network


Authors:
Yongqin Tao
Abstract
To deal with the evolutionary of complex network, a new evolutionary method using granular neural network (GNN) is proposed based on Yin Yang methodology. The theory of the granular quotient space is introduced to the neural network. At first input variables of the neural network are granulated to equivalence classes, so that the input variables of the network structure can be simplified and have certain clustering characteristics and strong diversity. And then the network parameters and the weights are optimized using evolutionary algorithms. Not only can the algorithm converge to globally optimal solution, but also it solves premature convergence problem efficiently. The simulation results show that the algorithm effectively narrows the search space and accelerates the speed of convergence.
Keywords
Complex Network; Granular Computing; Genetic Algorithm; Neural Network
StartPage
195
EndPage
203
Doi
Download | Back to Issue| Archive