OPTIMIZATION OF NEURAL NETWORKS BY GENETIC ALGORITHM
Iu L · Computer Applications and Software · 1997
A geneiic algorithm to optimize both neural network and related connection weights in a given task is proposed. A neural network having arbitrary connections is regarded as a virtural living thing which has genes respresenting ita connections among neural units. An XOR problem ia dealt with. The proposed method is available for designing a neural network whose adequate structure ia unknown.