Fast Conjugate Gradient Algorithm for Feedforward Neural Networks.

The conjugate gradient (CG) algorithm is a method for learning neural networks. The highest computational load in this method is directional minimization. In this paper a new modification of the conjugate gradient algorithm is presented. The proposed solution speeds up the directional minimization, which result in a significant reduction of the calculation time. This modification of the CG algorithm was tested on selected examples. The performance of our method and the classic CG method was compared.