Encoding and Identification of Finite State Machines via Recurrent Neural Networks
Sekon Won, Iickho Song, S.Y. Lee, Seungwon Lee, Yunsung Lee, K.Y. Kim · 2010
A new class of recurrent neural networks is proposed. The application of the proposed network is addressed in the encoding and identification of finite state machines (FSMs). Simulation results show that the identification of FSMs using the proposed network, trained by the hybrid greedy simulated annealing with a modified cost function in the learning stage, exhibits generally better performance than other conventional identification schemes.