Global Exponential Stability of High-Order Neural Networks with Time-Varying Coefficients and Delays

The paper presents a sufficient condition ensuring global exponential stability for high-order neural networks with time-varying coefficients and delays. The result allows for the consideration of all unbounded neuron activation functions, while the previous results allowed for the consideration of bounded activation functions. The method is based on basic analytical techniques and differential inequality techniques. The result of this paper is new and it complements previously known results. Several remarks are worked out to demonstrate the advantage of our result.