ANN模型及其学习算法，试图从数学上描述ANN的动力学过程，建立相应的ANN模型，在该模型的基础上，对于给定的学习样本，找出一种能以较快的速度和较高的精度调整神经元间互连权值，使系统达到稳定状态，满足学习要求的算法。-ANN model and its learning algorithm, trying to describe ANN from mathematics and the dynamics of the process, and set up a corresponding ANN model, in this model, and on the basis of learning samples for a given, to find a way to speed is fast and can adjust higher precision between neurons interconnection weights, make the system to achieve steady state, to meet the requirements of the learning algorithm.