Parallel radial basis function neural networks to solve the polynomials equations

The root-finding problem is one of the most important computational problems and applications. In this paper we introduced the modify artificial neural network is represented depending on radial basis function networks which have been three layers: input layer, hidden layer and output layer, where the hidden layer based on Gaussian function, these neural network techniques are developed to obtain the real approximate roots of single nonlinear equation with high accurate ratio. This modify RBFNN was used to proposed parallel environment for several intervals to compute approximate real roots.