| 1. | A general discussion for learning methods of fnns and an introduction to logic inference are firstly given . then , according to the analysis of networks and the evolution of their weights , a learning principle , based on superior contradiction reversion , is presented , which demonstrates the partition of the synaptic quality and the synaptic strength
首先讨论了前向神经网络的一般学习方法，介绍了神经网络定性逻辑推理的概念，然后分析了网络演化的依据和权值改变的规则，给出一种网络优势矛盾反传的学习原理，该原理体现了突触的性质与强度相区分的生物学事实。 |
