Runtime Analysis of OneMax Problem in Genetic Algorithm

Genetic algorithms (GAs) are stochastic optimization techniques, and we have studied the effects of stochastic fluctuation in the process of GA evolution. A mathematical study was carried out for GA on OneMax function within the framework of Markov chain model. We obtained the steady state solution, which represents a distribution of the first order schema frequency. We treated the task of estimating convergence time of the Markov chain for OneMax problem, and studied the effects of mutation probability and string length on the convergence time.