LEARNING AND EXTRACTING INITIAL MEALY AUTOMATA WITH A MODULAR NEURAL-NETWORK MODEL

A hybrid recurrent neural network is shown to learn small initial meal y machines (that can be thought of as translation machines translating input strings to corresponding output strings, as opposed to recognit ion automata that classify strings as either grammatical or nongrammat ical) from positive training samples. A well-trained neural net(1) is then presented once again with the training set and a Kohonen self-org anizing map with the ''star'' topology of neurons is used to quantize recurrent network state space into distinct regions representing corre sponding states of a mealy machine being learned. This enables us to e xtract the learned mealy machine from the trained recurrent network. O ne neural network (Kohonen self-organizing map) is used to extract mea ningful information from another network (recurrent neural network).