Neural network representation and identification of finite state automata
Yasuaki Kuroe, Taketoshi Mori · Society of Instrument and Control Engineers of Japan · 2003
This paper presents a new architecture of neural networks for representing deterministic finite state automata (FSA). The proposed model is capable of precisely representing FSA with the network size being smaller than the existing models proposed so far. We propose an identification method of FSA from a given set of input and output data by training the proposed neural networks.