慢速权值更新的ART2神经网络研究  

Research of algorithms of ART2 based on slow weight update rule

作　　者：叶晓明[1,2] 林小竹[1] 

机构地区：[1]北京石油化工学院信息工程学院,北京102617 [2]北京化工大学信息技术学院,北京100029

出　　处：《计算机工程与应用》2010年第24期146-150,共5页Computer Engineering and Applications

摘　　要：ART2是基于自适应谐振理论的一种自组织神经网络,通过竞争学习和自稳机制原理实现分类,可以在非平稳的、有干扰的环境中进行无监督的自学习,其学习过程能迅速识别已学习过的样本,并能迅速适应未学习过的新对象。提出了一种基于慢速权值更新的ART2神经网络算法,该算法在对输入模式进行识别分类时,会减慢学习速率,降低模式漂移的速度。新的网络学习规则在分类实验中取得了较好的效果,并在一定程度上解决了模式漂移问题。ART2 is a kind of self-organizing neural network which is based on adaptive resonance theory.It carries out the classification by using competitive learning and self-steady mechanism,and can learn by itself in dynamic environment with noise and without supervision.Its learning process can recognize learned models fast and be adapted to new unknown objects rapidly.This paper gives out an algorithm of ART2 based on slow weight update rule,which can slow down the learning rate and reduce the speed of pattern drifting in classification for input mode.This paper gets better results in experiment for classification by using the learning rule of new network,and to a certain extent solves the problem of pattern drifting.

关 键 词：自适应谐振理论 神经网络 模式漂移 分类 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]