摘要
基于牛顿向前插公式设计了一种新的联想记忆系(NFI-AMS)的学习算法,用以实现任意阶的多变量多项式函数的无误差逼近。该系统与传统类型的CMAC-AMS相比,具有学习精度高和记忆空间小的特点;且比多层的BP网络具有学习算法简单和收敛速度快的特点。数值模拟表明,这种NFI-AMS在信号处理,模式识别,及高精度的实时智能控制等领域具有很大的应用潜力。
A new high order Associative Memory System based on Newton’s Forward Interpolation formula (NFI AMS) used for implementing the error free approximation to multivariable polynomial functions with arbitrarily given order is proposed.The AMS posses the advantages over conventional CMAC type AMS in high precision of learning and much less required memory size and also the advantages over multi layer BP neural networks in much less computational effort for learning and fast convergence rate.Numerical simulations have shown that application areas of signal processing,pattern recognition,and controller implementation for high precision real time intelligent control. [
出处
《北京联合大学学报》
CAS
1998年第3期38-45,共8页
Journal of Beijing Union University
基金
国家自然科学基金
关键词
神经网络
联想记忆系统
学习算法
设计
neural network
associative memory system
function approximation
Newton's forward interpolation