A Novel Neural Network for Linear Complementarity Problems

Abstract:

      In this paper, we present a neural network for solving linear complementarity problem in real time. It possesses a very simple structure for implementation in hardware. In the theoretical aspect, this network is different from the existing networks which use the penalty functions or Lagrangians. We prove that the proposed neural network converges globally to the solution set of the problem starting from any initial point. In addition, the stability of the related differential equation system is analyzed and five numerical examples are given to verify the validity of the neural network.