Optimization of Neural Network Parameters Based on a Genetic Algorithm for Prediction of Time Series

The article proposes a method that allows to solve the complex combinatorial problem of structural optimization of an artificial neural network with a large dimension of the space of optimization parameters using the stochastic mutation function with a constant coefficient of variation. A study on assessment of applicability of this approach to solve the problem is given. This method is proposed to be used for the forecasting of time series values.