一种简单神经Mealy机的构造

构造一种新型神经Mealy机,神经Mealy机具有一定的学习能力,它主要通过学习来获得(von Newman)计算机结构,可以较好地避免普通计算机那样损毁一条电路就带来灾难性后果的情况.其本质是将递归神经网络通过BP优化算法,对Mealy机进行模拟得到,并通过实验对该网络的学习性能进行研究分析.基于形式文法和自动机的等价性,用神经网络来实现文法推导.先采用神经网络对样本集进行学习,这些样本可由一个经典Mealy机生成,然后从训练完的神经网络提取出自动机.

The Structure of a Simple Neural Mealy Machine

Construct a new type of neural Mealy machines,the neural Mealy machine has a certain capacity for learning;mainly for computer architecture by learning,can be avoided the catastrophic cons equences.Essentially,with BP optimizing algorithm of recurrent neural network,it can be obtained after simulating Mealy machines,and the learning performance of the network is analyzed through experiments.Based on the equivalence of formal grammars and automata,a neural network is used to realize the derivation of grammar.First neural network learns the sample that may be produced by a classical Mealy machine,and then we get an automata from training the neural network.