A Fast Learning Algorithm of Global Convergence for BP-Neural Network
一种快速且全局收敛的BP神经网络学习算法

There are many successful applications of back-propagation (BP) for training multi-layer neural networks. However, it has many shortcomings. Learning often takes long time to converge, and it may fall into local minima. In this paper, a fastlearning algorithm of global convergence for BP neural network is presented. Furthermore, the convergence of the optimization algorithm is analyzed in detail. A simulation example shows that the proposed algorithm is more efficient and accurate than the standard BP method.