Identification of the Fitness Function using Neural Networks
Oleg Kuzenkov, Galina Kuzenkova
Abstract
In this article, we demonstrate how the artificial neural networks can be applied to identify the fitness function of living organisms. The mathematical framework underlying this study is based on the analysis of observed long-time population dynamics that allows us to introduce the ranking order of inherited elements reflecting their selective advantages and to use Taylor approximation for the fitness function estimation. The developed technique is applied to find fitness of diel vertical migrations of zooplankton in the ocean and lakes. To identify the fitness function we created an artificial neural network that provides recognition of its coefficients taking into account a few key characteristics of the aquatic environment.