Feedforward neural network training using Grey Wolf Optimizer

Grey Wolf Optimizer (GWO) is a metaheuristic optimization method inspired by grey wolves which is suitable for solving optimization problems. The GWO algorithm mimics the leadership hierarchy and hunting mechanism of grey wolves in nature. First, grey wolf optimizer is presented. Next, it is employed for training feedforward neural networks for two benchmark classification problems. Then, the performance of GWO is compared with that of back-propagation (BP) methods. Simulation results demonstrate the effectiveness of the GWO algorithm.

کلیدواژه ها:

Artificial neural network ، back-propagation ، grey wolf ، optimization