神经网络的学习问题研究

神经网络的学习能力与效率问题是神经网络研究的一个重要方向，该文基于正交变换提出一种网络正交学习算法，它具有学习速度快且能获得全局最优解的特点，并可有效地对学习过程中出现的异常情况进行求解，因而具有良好的普适性。同时对新样本的学习可在以前学习的基础上继续，使网络的学习具有循序渐进的特征，提高了学习效率。

Research of the Learning Problems of Neural Networks

The learning ability and efficiency is an important problem of neural network research. This paper presents a neural network learning algorithm based on orthogonal transform, it has the characteristics of learning fast and getting the global best solution,and it can obtain the solution when ill condition occurs, so it also has good generalization. The new algorithm can learn following the previous learning, it can improve your learning efficiency, thus you can learn step by step through network.