论文部分内容阅读
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.
Constructing a new neuro Mealy machine, neuro Mealy machine has a certain learning ability. It mainly obtains the von Newman computer structure through learning, and can avoid the disastrous consequences caused by the common computer that damaging a circuit. Its essence is that the recurrent neural network is optimized by BP algorithm and the Mealy machine is simulated, and the learning performance of the network is studied through experiments.Graphically deduced by neural network based on the equivalence of formal grammar and automaton First, the neural network is used to study the sample set, which can be generated by a classic Mealy machine and then the automaton is extracted from the trained neural network.