Selection by parts: 'selection in two episodes' in evolutionary algorithms
Ambedkar Dukkipati, Musti Narasimha Murty · 2003
Naive models of evolution define natural selection as a process which brings in differential reproductive capabilities in organisms of a population, and hence, evolutionary algorithms implement selection by differential reproduction: the fittest members of the population are reproduced preferentially at the expense of the less fit members of the population. Formal models in evolutionary biology often subdivide selection into components, called 'episodes of selection', to capture the different complex mechanisms of nature by which Darwinian evolution can occur. In this paper we introduce the concept of 'episodes of selection' in evolutionary computation by means of a conceptual evolutionary model (ACE-model). This model captures selection in two episodes and in two phases of the evolutionary cycle. Here we give a formal description of the ACE-model, in which one can mechanize the two phases in different possible ways. We propose evolutionary algorithms based on the ACE-model, by giving simple mechanisms for implementation of two phases. Finally, we discuss the importance of introducing episodes of selection in evolutionary algorithms by simulations of the proposed evolutionary algorithms for function optimization.