The improvements of BP neural network learning algorithm

The back-propagation algorithm (BP) is a well-known method of training a multilayer feedforward artificial neural networks (FFANNS). Although the algorithm is successful, it has some disadvantages. Because of adopting the gradient method by the BP neural network, the problems including a slow learning convergent velocity and easily converging to local minimum can not be avoided. In addition, the selection of the learning factor and inertial factor affects the convergence of the BP neural network, which are usually determined by experience. Therefore the effective application of the BP neural network is limited. A new method in the BP algorithm to avoid a local minimum was proposed by means of adding gradually training data and hidden units. In addition, the paper also proposed a new model of a controllable feedforward neural network.