Markov chain analysis of genetic algorithms applied to fitness functions perturbed concurrently by additive and multiplicative noise

Abstract

We analyze the transition and convergence properties of genetic algorithms (GAs) applied to fitness functions perturbed concurrently by additive and multiplicative noise. Both additive noise and multiplicative noise are assumed to take on finitely many values. We explicitly construct a Markov chain that models the evolution of GAs in this noisy environment and analyze it to investigate the algorithms. Our analysis shows that this Markov chain is indecomposable; it has only one positive recurrent communication class. Using this property, we establish a condition that is both necessary and sufficient for GAs to eventually (i.e., as the number of iterations goes to infinity) find a globally optimal solution with probability 1. Similarly, we identify a condition that is both necessary and sufficient for the algorithms to eventually with probability 1 fail to find any globally optimal solution. Our analysis also shows that the chain has a stationary distribution that is also its steady-state distribution. Based on this property and the transition probabilities of the chain, we compute the exact probability that a GA is guaranteed to select a globally optimal solution upon completion of each iteration.

Multiobjective Lightning Flash Algorithm Design and Its Convergence Analysis via Martingale Theory

“…For any charge population state z subsequent to y, there is F(s t+1 � z) ≤ F(s t � y). en, it implies that z∈S F(z)P s t+1 � z s t � y ≤ z∈S F(y)P s t+1 � z s t � y � F(y) z∈S P s t+1 � z s t � y � F(y), (24) where the last equality holds due to S is a closed set, i.e., z∈S P(s t+1 � z | s t � y). Hence, we obtain that E F s t+1 s t � y ≤ F(y), (25) which indicates that F(s t ), t � 1, 2, .…”

“…Markov chain is a class of important stochastic processes with nonaftereffect property [17][18][19][20]. As it is proved that the optimization processes of some classical intelligent algorithms also have this property [21], Markov chain has been widely applied in convergence analysis of many intelligent algorithms [22][23][24][25][26][27][28], e.g., chicken swarm optimization, ant colony algorithm, and genetic algorithm (GA). In [22], the finite homogeneous Markov chain model of chicken swarm optimization is established with some of its properties being analyzed.…”

“…In [23], Huang et al analyze the convergence time of the ant colony algorithm based on the absorbing Markov chain model. Markov models of genetic algorithm and artificial bee colony algorithm are established, respectively, in [24,25]. e Markov chain model and a dynamic matrix model of the bat algorithm are proposed in [26].…”

The quasispecies regime for the simple genetic algorithm with ranking selection

“…It has been noted numerous times in the literature that a genetic algorithm is conveniently modelled as a Markov chain. Several researchers have studied genetic algorithms in this context, here is a selection of works belonging to this line of research: [2,4,16,25,33,34,35,37,38,39,51,52,59,61,53]. Unfortunately, the transition matrix is very complicated and the resulting formulas are intractable.…”