Noise-to-state practical stability and stabilization of random neural networks

This paper is devoted to studying noise-to-state practical stability and stabilization problems for random neural networks in the presence of general disturbances. It is proved that the existence and uniqueness of solutions is ensured if the noise intensity function is locally Lipschitz. Using random Lyapunov theory and the existence of practical Lyapunov functions, criteria are established for noise-to-state practical stability in mean of random neural networks. Some easily checkable and computable conditions are provided based on the structure characterization of the neural networks. Numerical examples are given to demonstrate the effectiveness of the developed methods.