Evolutionary Dynamics Analysis Using Complex Networks

Abstract

This thesis describes actual state of the selected evolutionary algorithm. It is focused on the last knowledge in the area of the differential evolution (DE) and it introduces a new approach to interactions between individuals in a population which are considered to be elements of a network. The idea to model evolutionary algorithm dynamics using network originates from the requirement to better understand relationships between individuals in a population and analyze the contribution of individuals in the process of the population improvement. The second part of this thesis presents novel algorithms developed on the basis on the previous analysis which encourage the relevance of this research.