Solving Graph Coloring Problems Based on a Chaos Neural Network with Non-monotonous Activation Function

The graph coloring problems is one of the classical combinatorial optimization problems having widespread applications in areas such as frequency assignment problems and computer compiler optimization. In this paper, a transiently chaotic neural network model with non-monotonous activation function for solving the graph coloring problem has been presented. By using hysteretic activation function which is multi-valued, adaptive, and has memory, the proposed model has higher ability of overcoming drawbacks that suffered from the local minimum and converge to the optimal solution quickly. From the numerical simulation results, it can be concluded that the proposed model has higher ability to search for globally optimal and has higher searching efficiency in solving the graph coloring problem.