An augmented neural network algorithm for solving singular convex optimization with nonnegative variables

Singular nonlinear convex optimization problems have been received much attention in recent years. Most existing approaches are in the nature of iteration, which is time-consuming and ineffective. Different approaches to deal with such problems are promising. In this paper, a novel neural network model for solving singular nonlinear convex optimization problems is proposed. By using LaSalle's invariance principle, it is shown that the proposed network is convergent which guarantees the effectiveness of the proposed model for solving singular nonlinear optimization problems. Numerical simulation further verified the effectiveness of the proposed neural network model.