Global Asymptotic Stability for Matrix-Valued Recurrent Neural Networks with Time Delays

This paper introduces matrix-valued recurrent neural networks with time delays, and proves the existence and uniqueness of the global equilibrium point. These networks are a generalization of complex-, quaternion- and Clifford-valued neural networks with matrix states. Two sufficient criteria are derived in terms of linear matrix inequalities that ensure the global asymptotic stability of the equilibrium point for the proposed networks. Finally, two simulation examples demonstrate the effectiveness of the theoretical results.