Michael D. Vose

Modeling genetic algorithms with Markov chains

We model a simple genetic algorithm as a Markov chain. Our method is both complete (selection, mutation, and crossover are incorporated into an explicitly given transition matrix) and exact; no special assumptions are made which restrict populations or population trajectories. We also consider the asymptotics of the steady state distributions as population size increases.

PATTERNS OF PARAPATRIC SPECIATION

Abstract. Geographic variation may ultimately lead to the splitting of a subdivided population into reproductively isolated units in spite of migration. Here, we consider how the waiting time until the first split and its location depend on different evolutionary factors including mutation, migration, random genetic drift, genetic architecture, and the geometric structure of the habitat. We perform large‐scale, individual‐based simulations using a simple model of reproductive isolation based on a classical view that reproductive isolation evolves as a by‐product of genetic divergence. We show that rapid parapatric speciation on the time scale of a few hundred to a few thousand generations is plausible even when neighboring subpopulations exchange several individuals each generation. Divergent selection for local adaptation is not required for rapid speciation. Our results substantiates the claims that species with smaller range sizes (which are characterized by smaller local densities and reduced dispersal ability) should have higher speciation rates. If mutation rate is small, local abundances are low, or substantial genetic changes are required for reproductive isolation, then central populations should be the place where most splits take place. With high mutation rates, high local densities, or with moderate genetic changes sufficient for reproductive isolation, speciation events are expected to involve mainly peripheral populations.

Representational issues in genetic optimization

Abstract Functions are partially characterized as easy or hard for genetic algorithms to optimize. The failure modes of inappropriate embedding, crossover disruption, and deceptiveness are introduced, analyzed, and resolved in part. Virtually all optimizable (by any method) real valued functions defined on a finite domain are shown to be theoretically easy for genetic algorithms given appropriately chosen representations. Unfortunately, problems that are easy in theory can be difficult in practice because of sampling error. Also, the transformations required to induce favorable representations are generally arbitrary permutations, and the space of permutations is so large that search for good ones is intractable. The space of inversions is amenable to search, but inversions are insufficiently powerful to overcome deceptiveness. On the other hand, affine transformations (over the diadic group) are shown to be sufficiently powerful to transform at least selected deceptive problems into easy ones. These new ...

Modeling simple genetic algorithms

The infinite- and finite-population models of the simple genetic algorithm are extended and unified, The result incorporates both transient and asymptotic GA behavior. This leads to an interpretation of genetic search that partially explains population trajectories. In particular, the asymptotic behavior of the large-population simple genetic algorithm is analyzed.

Generalizing the notion of schema in genetic algorithms

Abstract In this paper we examine some of the fundamental assumptions which are frequently used to explain the practical success which Genetic Algorithms (GAs) have enjoyed. Specifically, the concept of schema and the Schema Theorem are interpreted from a new perspective. This allows GAs to be regarded as a constrained random walk, and offers a view which is amenable to generalization. The minimal deceptive problem (a problem designed to mislead the genetic paradigm) is analyzed in the context provided by our interpretation, where a different aspect of its difficulty emerges.

RAPID PARAPATRIC SPECIATION ON HOLEY ADAPTIVE LANDSCAPES

A classical view of speciation is that reproductive isolation arises as a by–product of genetic divergence. Here, individual–based simulations are used to evaluate whether the mechanisms implied by this view may result in rapid speciation if the only source of genetic divergence are mutation and random genetic drift. Distinctive features of the simulations are the consideration of the complete process of speciation (from initiation until completion), and of a large number of loci, which was only one order of magnitude smaller than that of bacteria. It is demonstrated that rapid speciation on the time–scale of hundreds of generations is plausible without the need for extreme founder events, complete geographic isolation, the existence of distinct adaptive peaks or selection for local adaptation. The plausibility of speciation is enhanced by population subdivision. Simultaneous emergence of more than two new species from a subdivided population is highly probable. Numerical examples relevant to the theory of centrifugal speciation and to the conjectures about the fate of ‘ring species’ and ‘sexual continuums’ are presented.

A linear algorithm for generating random numbers with a given distribution

Let xi be a random variable over a finite set with an arbitrary probability distribution. Improvements to a fast method of generating sample values for xi in constant time are suggested. The proposed modification reduces the time required for initialization to O(n). For a simple genetic algorithm, this improvement changes an O(g n 1n n) algorithm into an O(g n) algorithm (where g is the number of generations, and n is the population size). >

GENETIC DIFFERENTIATION BY SEXUAL CONFLICT

Abstract Sexual conflict has been suggested as a general cause of genetic diversification in reproductive characters, and as a possible cause of speciation. We use individual-based simulations to study the dynamics of sexual conflict in an isolated diploid population with no spatial structure. To explore the effects of genetic details, we consider two different types of interlocus interaction between female and male traits, and three different types of intra-locus interaction. In the simulations, sexual conflict resulted in at least the following five regimes: (1) continuous coevolutionary chase, (2) evolution toward an equilibrium, (3) cyclic coevolution, (4) extensive genetic differentiation in female traits/genes only, and (5) extensive genetic differentiation in both male and female traits/genes. Genetic differentiation was hardly observed when the traits involved in reproduction were determined additively and interacted in a trait-by-trait way. When the traits interacted in a component-by-component way, genetic differentiation was frequently observed under relatively broad conditions. The likelihood of genetic differentiation largely depended on the number of loci and the type of within-locus dominance. With multiple loci per trait, genetic differentiation was often observed but sympatric speciation was typically hindered by recombination. Sympatric speciation was possible but only under restrictive conditions. Our simulations also highlight the importance of stochastic effects in the dynamics of sexual conflict.

Simple genetic algorithms with linear fitness

A general form of stochastic search is described (random heuristic search), and some of its general properties are proved. This provides a framework in which the simple genetic algorithm (SGA) is a special case. The framework is used to illuminate relationships between seemingly different probabilistic perspectives of SGA behavior. Next, the SGA is formalized as an instance of random heuristic search. The formalization then used to show expected population fitness is a Lyapunov function in the infinite population model when mutation is zero and fitness is linear. In particular, the infinite population algorithm must converge, and average population fitness increases from one generation to the next. The consequence for a finite population SGA is that the expected population fitness increases from one generation to the next. Moreover, the only stable fixed point of the expected next population operator corresponds to the population consisting entirely of the optimal string. This result is then extended by way of a perturbation argument to allow nonzero mutation.

The simple genetic algorithm and the walsh transform: Part i, theory

This paper is the first part of a two-part series. It proves a number of direct relationships between the Fourier transform and the simple genetic algorithm. (For a binary representation, the Walsh transform is the Fourier transform.) The results are of a theoretical nature and are based on the analysis of mutation and crossover. The Fourier transform of the mixing matrix is shown to be sparse. An explicit formula is given for the spectrum of the differential of the mixing transformation. By using the Fourier representation and the fast Fourier transform, one generation of the infinite population simple genetic algorithm can be computed in time O(cl log2 3), where c is arity of the alphabet and l is the string length. This is in contrast to the time of O(c3l) for the algorithm as represented in the standard basis. There are two orthogonal decompositions of population space that are invariant under mixing. The sequel to this paper will apply the basic theoretical results obtained here to inverse problems and asymptotic behavior.