The Application of Genetic Algorithm on the Training of Neural Network for Acoustic Target Classification

The paper adopted back-propagation neural network to classify acoustic target the wheeled and tracked vehicles was the researched target of this paper. Genetic Algorithm (GA) was first used to make global search of the suitable combination of the number of hidden nodes, the learning rate and momentum coefficient, the experiment in this paper will show that the neural network trained by GA has better performance in classifying wheeled and tracked target.