Theoretical studies of evolutionary algorithms (EAs) have existed since the seventies when EAs started to become popular. These early studies contributed towards the understanding of EAs but did not explain their performance in terms of their dynamical or limit behaviour. Only in the nineties the first convergence and time complexity results appeared for simple algorithms and toy problems. Nowadays, it is possible to analyse the time complexity of more complicated algorithms for combinatorial optimization problems with practical applications. This chapter overviews the most popular mathematical techniques used in the runtime analysis of EAs and gives some simple examples of their application. © 2011 by World Scientific Publishing Co. Pte. Ltd.
|Series on Theoretical Computer Science
|World Scientific Publishing Co.