A comparison of feed-forward back-propagation and radial basis artificial neural networks: A Monte Carlo study

Interest in soft computing techniques, such as artificial neural networks (ANN) is growing rapidly. Feed-forward back-propagation and radial basis ANN are the most often used applications in this regard. They have been utilized to solve a number of real problems, although they gained a wide use, however the challenge remains to select the best of them in term of accuracy and efficiency performance. This paper presents a comparison between feed-forward back-propagation and radial basis ANN base on their performance. The comparison is performed using a Monte Carlo study that involves the following problems: addition, multiplication, division, powers and a production function. The result indicates that the proposed radial basis ANN results are significantly better than proposed feed-forward back-propagation ANN results for all five problems.