Information Technologies of Evolutionarily Stable Behavior Recognition

Abstract

A software package was developed to find the quantitative and qualitative characteristics of evolutionarily stable vertical migrations of aquatic organisms according to the observed dynamical characteristics of the environment. The mathematical basis for solving the problem of predicting evolutionarily stable behavior is the maximum fitness principle. The method of deriving this function relies on the study of the population density dynamics over the space of hereditary elements. Evolutionally stable behavior is found by solving the classical problem of the calculus of variations, in which the fitness function is taken as the target functional. The software is built in the form of two interconnected complexes. The first complex provides a numerical solution of the problem of the calculus of variations to maximize the analytically defined fitness function of aquatic organisms under certain environmental conditions. This allows us to find evolutionarily stable behavior for a number of characteristic cases as the solution of the optimization problem. As a result, a training set of comparison samples is formed. It contains known precedents of the dynamic states of the environment and the corresponding strategies of evolutionarily stable vertical migrations of aquatic organisms. It is used for the recognition process. The second subsystem recognizes the qualitative characteristics of evolutionarily stable vertical migrations according to approximately given characteristics of the environment on the basis of using the formed base of the comparison samples.