Showing papers on "Selection (genetic algorithm) published in 1976"

The Algorithm Selection Problem

Abstract: Publisher Summary The problem of selecting an effective algorithm arises in a wide variety of situations. This chapter starts with a discussion on abstract models: the basic model and associated problems, the model with selection based on features, and the model with variable performance criteria. One objective of this chapter is to explore the applicability of the approximation theory to the algorithm selection problem. There is an intimate relationship here and that the approximation theory forms an appropriate base upon which to develop a theory of algorithm selection methods. The approximation theory currently lacks much of the necessary machinery for the algorithm selection problem. There is a need to develop new results and apply known techniques to these new circumstances. The final pages of this chapter form a sort of appendix, which lists 15 specific open problems and questions in this area. There is a close relationship between the algorithm selection problem and the general optimization theory. This is not surprising since the approximation problem is a special form of the optimization problem. Most realistic algorithm selection problems are of moderate to high dimensionality and thus one should expect them to be quite complex. One consequence of this is that most straightforward approaches (even well-conceived ones) are likely to lead to enormous computations for the best selection. The single most important part of the solution of a selection problem is the appropriate choice of the form for selection mapping. It is here that theories give the least guidance and that the art of problem solving is most crucial.

Genetic Polymorphism in Heterogeneous Environments

Abstract: The discovery in the last decade, using electrophoresis, of large amounts of genetic polymorphism in natural populations has had a tremendous effect on population genetics. Before then it was not clear how many polymorphic loci there were in a population and the single gene overdominance model generally seemed adequate to explain what polymorphism was documented. In an effort to explain the new-found genetic variation, two opposing camps surfaced in the late 1960s and still exist to some extent today. One group, often called neutralists, believes that most allozymic variants have a minimal effect on fitness and are in a population because of a combination of mutation, finite population size, and migration. In other words, in their view, selection plays little or no role in maintaining different electrophoretic alleles. The other group, sometimes known as selectionists, believes that some sort of balancing selection is responsible for the maintenance of the majority of electropho­ retic alleles. "Balancing selection" is a catch all term for any type of selection that can maintain a stable polymorphism. Among the types of balancing selection that are thought to play a significant role in maintaining genetic polymorphisms are the classical overdominance mode, frequency-dependent selection [see (8) for a recent review], differential selection between the two sexes or between different life stages, and variable selection in time and/or space. This last type of balancing selection is examined here in an effort to understand the importance of selection varying in time and/or space in maintaining genetic polymQrphisms. A significant body of literature demonstrates an association between the genetic attributes of a population and some aspect of the environment. Since much of this material has been reviewed elsewhere (2, 16,58, 165), we concentrate on two of the

The theoretical population genetics of variable selection and migration

Abstract: Selection varying with time is reviewed with regard to infinite and finite populations. Geographic effects are reviewed with regard to migration versus selection; genetic drift and migration; and selection. Various models of geographic differentiation are described. (HLW)

Why be an hermaphrodite

Abstract: MANY animals and most higher plants are hermaphrodites. The basic argument of this paper is that the sex habit of a species is determined by selection acting on the number of offspring produced by individuals of different types. The argument differs radically from most earlier explanations of the evolution of hermaphroditism (reviewed by Ghiselin)1,2, although it is formally similar to a recent explanation3 of sequential hermaphroditism, in which individuals function first as one sex and then the other.

Group SelectionGroup Selection in Predator-Prey Communities. M. E. Gilpin

Abstract: T H HE PURPOSE of this short review is to look at some recent discussions of group selection, in particular those by E. 0. Wilson (1975), D. S. Wilson (1975) and M. E. Gilpin (1975). Earlier work will be referred to only briefly; the need for a review arises because the three references given either propose a blurring of the distinction between \"group\" and \"kin\" selection, or suggest an importance for group selection greater than it has usually been given, or both. The first point to establish is that the argument is quantitative, not qualitative. Group selection will have evolutionary consequences; the only question is how important these consequences have been. If there are genes which, although decreasing individual fitness, make it less likely that a group (deme or species) will go extinct, then group extinction will influence evolution. It does not follow that the influence is important enough to play the role suggested for it by some biologists. The present phase of the debate about group selection was opened by the publication of Wynne-Edwards' \"Animal Dispersion\" (1962), which applied to animals a concept first proposed by Carr-Saunders (1922) to explain human population dynamics; it ascribed to group selection a major role in the evolution of population regulation. Although his thesis has had its adherents, the orthodox response from ecologists has been to argue that the patterns of behavior he described can be explained by individual selection, and from population geneticists that the mechanism he proposes is insufficient to account for the results. It is in the nature of science that once a position becomes orthodox it should be subjected to criticism; hence the papers by D. S. Wilson (1975), Gilpin (1975), Levin and Kilmer (1974), Gadgil (1975), and others. It does not follow that, because a position is orthodox, it is wrong; hence this review. Is the argument important? In a recent review of E. 0. Wilson's \"Sociobiology,\" C. H. Waddington referred to group selection as \"a fashionable topic for a rather foolish controversy.\" Doubtless some foolish things have been said, but there is an important issue at stake. If group selection has played the role suggested by Wynne-Edwards, no one can doubt its importance. But why should it be important to argue that it has not? The reason for the vehemence with which Williams (1966, 1975), Ghiselin (1974), Lack (1966) and other opponents of group selection have argued their case is, I think, their conviction that group selection assumptions, often tacit or unconscious, have been responsible for the failure to tackle important problems. So long as we fail to distinguish group and individual selection, or assume that an explanation in terms of advantage to the species is adequate to account for the evolution of some behavior pattern or genetic process, without asking what is its effect on individual fitness, we shall make little progress. The extent of unconscious group selection-

Analysis of Tetrapod Gaits: General Considerations and Symmetrical Gaits

Abstract: The relative timing of the cyclic contacts that the feet of tetrapods make with the ground in terrestrial locomotion determines the gaits of the animals. This chapter reports on a comprehensive and integrative study that establishes a system for analyzing gaits. The model facilitates description, identifies all possible gaits, permits the simultaneous study of hundreds of locomotor performances, and helps to interpret the selection of gaits by the various animals.

An evaluation of some models for culture-fair selection

Abstract: in this process. There are many different definitions of what constitutes culture-fair selection, and each implicitly, though unfortunately not explicitly, involves a particular set of value judgments with different implications for how selection should be accomplished. For each of these definitions a remedy has been proposed. Our purpose in this paper is to show that some of these approaches are inadequate to their task and that more complex analyses are required. DESCRIPTION OF THE SELECTION PROCESS The selection process can be characterized in the same manner for all selection models. First, there are individuals about whom decisions are required. These decisions are to be based on information about the individuals. The information is processed by some strategy which leads to a final decision. The final decision ends the decision-making process by assigning the individual to either a selected or an outselected group. The outcome is the individual's performance after the assignment or, in other words, the consequences resulting from the decision (Cronbach & Gleser, 1965, p. 18). A strategy is a rule for making decisions. Each selection model represents a strategy, the intent of which is to guarantee cultural fairness in the selection process. Information is generally provided by a test, and we shall use the term test to refer to all information-gathering procedures including interviews and physical measurements. The overriding problem is the lack of agreement as to the meaning of the term culture-fair selection. Each of the selection models or strategies we discuss can be characterized in the same manner. It is assumed that the applicants to an educational institution, to a training program, or for employment can be separated into subpopulations because of an a priori belief that the assumed linear regressions within these subpopulations are different-that is, the test (or predictor) may be more valid for some subpopulations than for others (different slopes), and/or for a fixed value of the predictor, the level of criterion performances may differ (different intercepts), or that some differential selection criterion is appropriate for various subpopulations. Alternatively, these subpopulations may be differentiable primarily because of public concern with what is going on in them and a public need to verify that all subpopula

It's the Thought that Counts: A Signed Digraph Analysis of Gift-Giving

Abstract: This article presents and tests a model of gift selection based on cognitive consistency theories. The model, which is supported by the data presented, offers a means for predicting the conditions under which giver tastes or recipient tastes dominate gift selection, and for predicting the amount of satisfaction which the gift brings the giver.

Group selections among laboratory populations of Tribolium.

Abstract: Selection at the population level or group selection is defined as genetic change that is brought about or maintained by the differential extinction and/or proliferation of populations. Group selection for both increased and decreased adult population size was carried out among laboratory populations of Tribolium castaneum at 37-day intervals. The effect of individual selection within populations on adult population size was evaluated in an additional control series of populations. The response in the group selection treatments occurred rapidly, within three or four generations, and was large in magnitude, at times differing from the controls by over 200%. This response to selection at the populational level occurred despite strong individual selection which caused a decline in the mean size of the control populations from over 200 adults to near 50 adults in nine 37-day intervals. "Assay" experiments indicated that selective changes in fecundity, developmental time, body weight, and cannibalism rates were responsible in part for the observed treatment differences in adult population size. These findings have implications in terms of speciation in organisms whose range is composed of many partially isolated local populations.

Niche Width in a Fluctuating Environment-Density Independent Model

Abstract: A model of the effect of density-independent optimizing selection on the distribution of a character is developed. Both the direct effect of the selection on the character and the secondary effect of the selection on a modifier allele which changes the distribution is analyzed. The results are that there would be selection for a lower variance in a constant environment and selection for a larger variance when the optimal value of the character is changing in time, only if the fluctuations in the optimum exceed a certain threshold amplitude. The application of the model to populations subject to different temporal patterns of fluctuating selection is discussed.

Objectives in pig breeding

Abstract: 1. Some criticisms of the classical approach to constructing a selection index are advanced. This approach (the economic model) is contrasted with an alternative one (the biological model) which is derived from the concept of the biological efficiency of production of lean tissues. The selection objective which is proposed from the biological model is the improvement of lean tissue feed conversion. The most important single means of achieving this is increasing the lean tissue growth rate.2. The approaches were compared in three studies: (a) the effects of errors in economic and genetic parameters of the economic index; (b) phenotypic relationships between biological objectives and other measures of performance; (c) a theoretical model based on the utilization of metabolizable energy.3. In a fixed situation the results were similar for both models, but the main advantage of the biological model is that it gives good indications of what happens, and what selection policies should be, in different situations. In addition, lean tissue feed conversion and lean tissue growth rate are closely related to concepts widely applied in biology, and their use as selection objectives should be helpful in enabling workers in many disciplines to make use of each other's knowledge.

Two-way selection for growth rate in the common carp (cyprinus carpio l.)

Abstract: The domesticated European carp was subjected to a two-way selection for growth rate. Five generations of mass selection for faster growth rate did not yield any response, but subsequent selection between groups (families) resulted in considerable progress while maintaining a large genetic variance. Selection for slow growth rate yielded relatively strong response for the first three generations. Random-bred control lines suffered from strong inbreeding depression and when two lines were crossed, the F1 showed a high degree of heterosis. Selection was performed on pond-raised fish, but growth rate was also tested in cages. A strong pond-cage genetic interaction was found. A theoretical explanation was suggested involving overdominance for fast growth rate and amplification through competition of intra-group but not inter-group variation.

An applied model for the facilities design problem

Abstract: A model ia described which can be used to assist in the design of a facility, where the facilities design includes the selection of the materials handling system and the placement of departments within the facility. The model is entitled COFAD, an acronym representing Computerized FAcilities Design. COFAD selects the facilities design which approaches the minimal materials handling system cost. The input requirements and the utilization of COFAD are demonstrated via an example problem. Although it is shown that the model results in good solutions, due to the complexity of the design problem, no claim of optimality can be made.

Laboratory Tests for the Assessment of Nutritional Status

Abstract: This book is a timely review of the laboratory methods currently available and provides a recognition of their usefulness and limitations in evaluating nutritional status of human population. Literature citations in the text provide guidance for the selection of precise methods.

The effect of selection on genetic variability: a simulation study.

Abstract: Computer simulations have been done to study the effects of stabilizing and disruptive selection on a polygenic character. The results are reported in terms of three components of genetic variability which represent changes in gene frequencies, departures from Hardy–Weinberg equilibrium and linkage disequilibrium respectively. Under random mating the first and third components are the most important. The observed changes in gene frequencies are interpreted in the light of previous theoretical work on the stability of equilibria under selection. In addition, large and rapid changes in the genotypic variance result from the generation of linkage disequilibrium under selection; the observed changes are in good agreement with those predicted on theoretical grounds.

Model Search among Multiplicative Models

Abstract: We propose a non-iterative model search technique to find simple patterns of association for several variables. Our selection procedure is restricted to multiplicative models, therefore all patterns under consideration are interpretable in terms of zero partial associations of variable pairs. Wie illustrate the selection technique on two sets of data, one in a contingency table, one in a covariance matrix.

The Effect of Individual and Situation-Related Factors on Consumer Selection of Judgmental Models:

Abstract: The study reported tests the hypotheses that the consumer's use of a specific judgmental model in evaluating a product is a function of prior familiarity and/or product complexity. The results indi...

Order statistics of correlated variables and implications in genetic selection programmes.

Abstract: Formulae are derived and examples given for the expected values of the order statistics from a sample of several groups or families of equal size where the group members have intra-class correlation t. As t increases from O the means of the highest ranking individuals are little reduced initially, but as t approaches 1 the change becomes more rapid, the total reduction depending on the number of groups. The effect on selection differentials of the intra-class correlation of family members is generally small for t less than 0.5, such as if selection is practised on individual performance. If, however, family mean performance is used in an index, the selection differentials may be substantially reduced.

A model of competition for food. i. frequency- dependent viabilities

Abstract: A model is developed in which animals compete for a limited amount of food. The principal parameters are individual properties, the minimum amount of food necessary for survival, m, and the probability of acquiring a food particle per time unit, P. The probability of survival is the probability of acquiring the minimum amount of food or more, the right tail of the distribution of food particles. Although in this model individuals interact only by their exploitation of the same resource and no interference or facilitation is present, depending on the parameters both frequency-dependent and frequency-independent '"selection pertain. Selection always increases "efficiency." Viability in pure culture has no predictive value for viability in mixed culture; the latter can be predicted if m and P are known.

Bank Selection Decisions and Market Segmentation: Determinant attribute analysis reveals convenience- and sevice-oriented bank customers

Abstract: I N an analysis of marketing activities in service industries, George and Barksdale concluded that although service industries have experienced unprecedented expansion in recent years, \"the marketing function appears to be less structured in service_companies than in manufacturing firms.... Fragmentation of marketing activities in service firms holds true for all components of the marketing mix.\"! Moreover, service firms appear to allocate a relatively smaller proportion of their operating budgets to marketing activity than manufacturing firms. Marketing has traditionally been the neglected stepsister in most service industries. Nowhere has this been more evident than in the market for commercial banking services.? Banking has historically had an apparent immunity from marketing. Yet,