Feedforward Neural Network Based on Improved Gray Wolf Optimizer

Neural network is one of the greatest inventions in the field of artificial intelligence. It mimics the neural elements of the human brain and is mainly used to solve classification problems and make data prediction, parameter optimization of which remains to be an important problem. Different results are obtained by different algorithms. In the paper, the gray wolf optimizer algorithm based on bionic simulation is studied. By optimizing the non-linear convergence factor and making it more consistent with the actual convergence process of the algorithm, so it can better balance the global search performance with the local search performance, and further enhance the global optimization ability of the algorithm. By increasing the dynamic weight, the gray wolf of the leadership can dynamically guide the gray wolf herd forward, which greatly enhances the adaptability of the algorithm to the environment. The experimental results show that the performance of the optimized feedforward neural network based on Grey Wolf Optimizer(GWO) algorithm performs better than that of the previous ones.