A robust complex network generation method based on neural networks

To enhance the network tolerance against numerous network attack strategies, various techniques to optimize conventional complex networks, such as scale-free networks, have been proposed. In this paper, we propose a new optimization technique based on artificial neural networks that is trained on scale-free network topologies as input data and hill climbing network topologies as output data. The goal of our method is to provide similar network robustness as the hill climbing network with much reduced complexity. Based on the experimental results, we demonstrate that the proposed network can provide strong robustness against both random and targeted attack, while significantly reduce optimization complexity.