Learning regular languages by neural networks

In this paper, we introduce a new method for learning regular languages using neural networks. The method is based on two linear time algorithms and assume that the size of the alphabet is constant. The first algorithm constructs the initial neural network from the maximal length (i.e. M) of the words, belonging in a positive sample, and the site of the alphabet. The second algorithm (i.e. the learning algorithm) takes the sample, the value of M, and the initial neural network as inputs, and constructs by adjusting the weights of the final neural network.