Showing papers on "Selection (genetic algorithm) published in 1996"
Proceedings Article•
03 Jul 1996
TL;DR: An efficient algorithm for feature selection which computes an approximation to the optimal feature selection criterion is given, showing that the algorithm effectively handles datasets with a very large number of features.
Abstract: In this paper, we examine a method for feature subset selection based on Information Theory. Initially, a framework for defining the theoretically optimal, but computationally intractable, method for feature subset selection is presented. We show that our goal should be to eliminate a feature if it gives us little or no additional information beyond that subsumed by the remaining features. In particular, this will be the case for both irrelevant and redundant features. We then give an efficient algorithm for feature selection which computes an approximation to the optimal feature selection criterion. The conditions under which the approximate algorithm is successful are examined. Empirical results are given on a number of data sets, showing that the algorithm effectively handles datasets with a very large number of features.
1,713 citations
22 Sep 1996
TL;DR: The problem is the problem and several modifications of sexual recombination are investigated, which leads to marginal distribution algorithms, which lead to more sophisticated methods, based on estimating the distribution of promising points.
Abstract: The Breeder Genetic Algorithm (BGA) is based on the equation for the response to selection. In order to use this equation for prediction, the variance of the fitness of the population has to be estimated. For the usual sexual recombination the computation can be difficult. In this paper we shortly state the problem and investigate several modifications of sexual recombination. The first method is gene pool recombination, which leads to marginal distribution algorithms. In the last part of the paper we discuss more sophisticated methods, based on estimating the distribution of promising points.
1,269 citations
TL;DR: It is concluded that a combination of directional and stabilizing selection is a prevalent mode of selection and a likely explanation for the evolution of modularity.
Abstract: Synopsis. The fact that phenotypic evolution can be studied on a char? acter by character basis suggests that the body is composed of locally integrated units. These units can be considered as modular parts of the body which integrate functionally related characters into units of evolu? tionary transformation. These units may either emerge spontaneously by self-organization, or may be the product of natural selection. A selection scenario that could explain the origin of modular units needs to explain the differential suppression of pleiotropic effects between different modules and the augmentation of pleiotropic effects among the elements with? in the module. Four scenarios are discussed: selection for adaptation rate, constructional selection, stabilizing selection and a combination of direc? tional and stabilizing selection. It is concluded that a combination of directional and stabilizing selection is a prevalent mode of selection and a likely explanation for the evolution of modularity.
727 citations
TL;DR: In this article, the authors outline the issues relevant to regression test selection techniques, and use these issues as the basis for a framework within which to evaluate the techniques and illustrate the application of the framework by using it to evaluate existing regression-test selection techniques.
Abstract: Regression testing is a necessary but expensive maintenance activity aimed at showing that code has not been adversely affected by changes. Regression test selection techniques reuse tests from an existing test suite to test a modified program. Many regression test selection techniques have been proposed, however, it is difficult to compare and evaluate these techniques because they have different goals. This paper outlines the issues relevant to regression test selection techniques, and uses these issues as the basis for a framework within which to evaluate the techniques. The paper illustrates the application of the framework by using it to evaluate existing regression test selection techniques. The evaluation reveals the strengths and weaknesses of existing techniques, and highlights some problems that future work in this area should address.
651 citations
TL;DR: The authors showed that the sample selection model is susceptible to collinearity problems and a t -test can be used to distinguish between the two models as long as there are no collinearlyity problems.
519 citations
TL;DR: A mathematical analysis of tournament selection, truncation selection, ranking selection, and exponential ranking selection is carried out that allows an exact prediction of the fitness values after selection.
Abstract: Evolutionary algorithms are a common probabilistic optimization method based on the model of natural evolution. One important operator in these algorithms is the selection scheme, for which in this paper a new description model, based on fitness distributions, is introduced. With this, a mathematical analysis of tournament selection, truncation selection, ranking selection, and exponential ranking selection is carried out that allows an exact prediction of the fitness values after selection. The correspondence of binary tournament selection and ranking selection in the expected fitness distribution is proved. Furthermore, several properties of selection schemes are derived (selection intensity, selection variance, loss of diversity), and the three selection schemes are compared using these properties.
460 citations
TL;DR: It is shown that crossing of reaction norms for fitness is a sufficient condition for selection to favor specialized host choice, in the absence of search costs and other disadvantages of specialization.
Abstract: The idea that traits that adapt a population to one habitat or host are deleterious in other habitats or on other hosts is central to most theories for the evolution of ecological specialization. The trade-off concept has played an especially prominent role in discussions of the evolution of host specialization in phytophagous insects. The evidence for genetically based trade-offs in phytophagous insect populations is ambiguous, however. Cross-host genetic corre- lations for fitness traits are seldom negative, but this does not preclude the existence of trade-offs at a subset of loci controlling the fitness variation. One clear result is that cross-host genetic correlations for fitness traits are often less than one, which implies that genotypes have different fitness rankings on different hosts (i.e., that reaction norms for fitness cross). Using verbal arguments and a mathematical model, I show that crossing of reaction norms for fitness is a sufficient condition for selection to favor specialized host choice, in the absence of search costs and other disadvantages of specialization. In other words, selection will favor specialization if alleles that are positively selected on one host are less strongly positively selected, or neutral, on other hosts; it is not necessary for the alleles to be deleterious on the other hosts. Search costs and other factors may oppose the evolution of specialization, but introducing trade-offs into a model does not result in a quantum jump in the strength of selection favoring, specializa- tion, and thus in the likelihood of specialization evolving, compared to the situation in which alleles that affect fitness on one host are neutral on others. There is therefore no justification for focusing on the qualitative presence or absence of trade-offs as the critical issue in predicting or explaining the evolution of specialization. Furthermore, the quantitative genetic data, rather than give little information on why specialization evolves, indicate that the potential for selection to favor specialization exists in many phytophagous populations.
377 citations
TL;DR: Models for several selection schemes are developed that successfully predict the convergence characteristics of GAs within noisy environments that include proportionate selection, tournament selection, (, ) selection, and linear ranking selection.
Abstract: This paper analyzes the effect of noise on different selection mechanisms for genetic algorithms (GAs). Models for several selection schemes are developed that successfully predict the convergence characteristics of GAs within noisy environments. The selection schemes modeled in this paper include proportionate selection, tournament selection, (μ, λ) selection, and linear ranking selection. An allele-wise model for convergence in the presence of noise is developed for the OneMax domain, and then extended to more complex domains where the building blocks are uniformly scaled. These models are shown to accurately predict the convergence rate of GAs for a wide range of noise levels.
373 citations
TL;DR: This article develops an integrated measure of fitness based on the philosophical underpinnings provided by the notion of propensity fitness, combined with classical demographic methods, and shows how life-history data can be formatted as an age-structured population projection matrix, A(m).
Abstract: Fitness is a vague, poorly developed concept in ecology and evolutionary biology. Although it is generally recognized that fitness is determined by the complete survival and reproductive schedules of individual organisms, experimental studies have rarely attempted to integrate these into a single measure of individual fitness. Instead, particularly in studies of natural selection, components of fitness are measured as surrogates for fitness. In this article, we develop an integrated measure of fitness based on the philosophical underpinnings provided by the notion of propensity fitness, combined with classical demographic methods. We show how life-history data, collected for an individual, can be formatted as an age-structured popula- tion projection matrix, A("?). The dominant eigenvalue, X('t), of this matrix is an estimate of that individual's propensity fitness. Using life-history data sets on European sparrowhawks and blue tits, we show that the interpretation of analyses of selection can shift radically using the inte- grated propensity fitness measure instead of components of fitness, such as lifetime reproductive output. Individual fitness, as formulated here, provides an integrated measure of performance that should prove useful in empirical studies in ecology and evolutionary biology.
358 citations
TL;DR: Critical factors that must be considered to identify likely plant attributes that can be targeted and which show substantial promise in targeting traits to improve yield under drought are discussed.
Abstract: The many selection criteria that have been proposed to increase drought resistance of our crops have had little, if any, impact on improving crop yields in dry environments. There are several likely reasons for this lack of success. Some of these are: (i) criteria proposed have been related more to survival mechanisms under drought than to productivity, (ii) criteria are inappropriate to the target environment, and (iii) criteria are temporal and are therefore likely to have minimal impact on growth and yield over the entire lifecycle. Another important reason is that breeders have not been convinced the proposed criteria will be successful as they are too difficult to measure. On the other hand, empirical breeding programmes to improve yield under drought have been successful. Surprisingly, some of the greatest successes have been achieved by breeding in environments where water is non-limiting. This paper reviews breeding approaches to improve yield under drought. It focuses on critical factors that must be considered to identify likely plant attributes that can be targeted. These factors, their link with yield, the nature of the target environment, the level of organisation where the trait is expressed are discussed. Three quite different examples are given to emphasize the above considerations and which show substantial promise in targeting traits to improve yield under drought. They are drought at flowering, improving transpiration efficiency and improving early leaf area development.
350 citations
Book•
01 Jan 1996
TL;DR: Tracer selection; data acquistion; data processing; kinetic analysis; statistical analysis; tracing; and data acquisition and processing.
Abstract: Tracer selection; data acquistion; data processing; kinetic analysis; statistical analysis. (Part contents).
TL;DR: It is argued here that selection procedures based on global information should be used, at at early stages of a test when θ estimates are not likely to be close to the true θ, and an item selection procedure based on average global information is proposed.
Abstract: Most item selection in computerized adaptive testing is based on Fisher information (or item information). At each stage, an item is selected to maximize the Fisher information at the currently estimated trait level (θ). However, this application of Fisher information could be much less efficient than assumed if the estimators are not close to the true θ, especially at early stages of an adaptive test when the test length (number of items) is too short to provide an accurate estimate for true θ. It is argued here that selection procedures based on global information should be used, at least at early stages of a test when θ estimates are not likely to be close to the true θ. For this purpose, an item selection procedure based on average global information is proposed. Re sults from pilot simulation studies comparing the usual maximum item information item selection with the pro posed global information approach are reported, indicat ing that the new method leads to improvement in terms of bias and mean squ...
TL;DR: Escherichia coli is used to provide an estimate of the genomic deleterious-mutation rate for total fitness in a microbe and it is estimated that the per-microbe rate of deleteriously mutations is in excess of 0.0002.
Abstract: Mutations are a double-edged sword: they are the ultimate source of genetic variation upon which evolution depends, yet most mutations affecting fitness (viability and reproductive success) appear to be harmful Deleterious mutations of small effect can escape natural selection, and should accumulate in small population Reduced fitness from deleterious-mutation accumulation may be important in the evolution of sex, mate choice, and diploid life-cycles, and in the extinction of small populations Few empirical data exist, however Minimum estimates of the genomic deleterious-mutation rate for viability in Drosophila melanogaster are surprisingly high, leading to the conjecture that the rate for total fitness could exceed 10 mutation per individual per generation Here we use Escherichia coli to provide an estimate of the genomic deleterious-mutation rate for total fitness in a microbe We estimate that the per-microbe rate of deleterious mutations is in excess of 00002
Book•
08 Oct 1996
TL;DR: This chapter discusses mating strategies based on Pedigree Relationship: Inbreeding and Outbreeding, and selection for Simply-Inherited Traits and Multiple-Trait Selection.
Abstract: I. ANIMAL BREEDING FROM THE TOP DOWN. 1. What is the "Best" Animal? 2. How are Animal Populations Improved? II. ANIMAL BREEDING FROM THE BOTTOM UP. 3. Mendelian Inheritance. 4. Genes in Populations. 5. Simply-Inherited and Polygenic Traits. III. SELECTION. 6. Selection for Simply-Inherited Traits. 7. The Genetic Model for Quantitative Traits. 8. Statistics and Their Application to Quantitative Traits. 9. Heritability and Repeatability. 10. Factors Affecting the Rate of Genetic Change. 11. Genetic Prediction. 12. Large-Scale Genetic Evaluation. 13. Correlated Response to Selection. 14. Multiple-Trait Selection. IV. MATING SYSTEMS. 15. Selection for Simply-Inherited Traits. 16. Mating Strategies Based on Animal Performance: Random and Assortative Mating. 17. Mating Strategies Based on Pedigree Relationship: Inbreeding and Outbreeding. 18. Hybrid Vigor. 19. Crossbreeding Systems. V. NEW TECHNIQUES, OLD STRATEGIES Glossary. Appendix. Answers to Odd-Numbered Problems. Index.
01 May 1996
TL;DR: The case study indicated that the evaluated aspects of the method are feasible, improve the quality and efficiency of reusable software selection and the decision makers have more confidence in the evaluation results, compared to traditional approaches.
Abstract: This paper describes a case study that used and evaluated key aspects of a method developed for systematic reusable off-the-shelf software selection. The paper presents a summary of the common problems in reusable off-the-shelf software selection, describes the method used and provides details about the case study carried out. The case study indicated that the evaluated aspects of the method are feasible, improve the quality and efficiency of reusable software selection and the decision makers have more confidence in the evaluation results, compared to traditional approaches. Furthermore, the case study also showed that the choice of evaluation data analysis method can influence the evaluation results.
TL;DR: This paper examined the bases for fairness reactions to different selection practices and considered cross-cultural differences in these reactions by comparing respondents from 2 cultures and found that the perceived face validity of the selection procedure was the strongest correlate of favorability reactions among both samples.
Abstract: The authors examined the bases for fairness reactions to different selection practices and considered cross-cultural differences in these reactions by comparing respondents from 2 cultures. College students (Ar = 259) from France and the United States rated the favorability of 10 selection procedures and then indicated the bases for these reactions on 7 procedural dimensions. Selection decisions based on interviews, work-sample tests, and resumes were perceived favorably in both cultures. Graphology was perceived more favorably in France than in the United States, but even French reactions toward graphology were somewhat negative. The perceived face validity of the selection procedure was the strongest correlate of favorability reactions among both samples. Beyond comparing the results from each culture, the discussion addresses implications for multinational companies establishing selection systems in foreign countries.
01 Jan 1996
TL;DR: It is shown that many features of the observed relations between chromosomal location and level of genetic diversity in D. melanogaster can be explained by background selection, especially if the weak selective forces acting on transposable elements are taken into account.
Abstract: Theoretical models of the effects of selection against deleterious mutations on variation at linked neutral sites (background selection) are used to predict the relations between chromosomal location and genetic variability at the DNA level, in Drosophila melanogaster. The sensitivity of the predictions to variation in the mutation, selection and recombination parameters on which they are based is examined. It is shown that many features of the observed relations between chromosomal location and level of genetic diversity in D. melanogaster can be explained by background selection, especially if the weak selective forces acting on transposable elements are taken into account. In particular, the gradient in diversity in the distal portion of the X chromosome, and the lack of diversity on chromosome 4 and at the bases of the major chromosomes, can be fully accounted for. There are, however, discrepancies between predicted and observed values for some loci in D. melanogaster, which may reflect the effects of forces other than background selection.
TL;DR: Results suggest that the good genes process generates only weak indirect selection on preferences, with an effective selection intensity of a few percent or less, and direct selection of preference genes may play an important role in species that show extreme sexual selection.
Abstract: A model is used to study quantitatively the impact of a good genes process and direct natural selection on the evolution of a mating preference. The expression of a male display trait is proportional to genetic quality, which is determined by the number of deleterious mutations a male carries throughout his genome. Genetic variances and covariances, including the covariance between the preference and male trait that drives the good genes process, are allowed to evolve under an infinitesimal model. Results suggest that the good genes process generates only weak indirect selection on preferences, with an effective selection intensity of a few percent or less. If preferences are subject to direct natural selection of the intensity observed for other characters, the good genes process alone is not expected to exaggerate the male trait by more than a few phenotypic standard deviations, contrary to what is observed in highly sexually selected species. Good genes can, however, cause substantial exaggeration if preference genes are nearly selectively neutral. Alternatively, direct selection on preference genes, acting on mating behavior itself or on the genes' pleiotropic effects, can cause mating preferences and male display traits to be exaggerated by any degree. Direct selection of preference genes may therefore play an important role in species that show extreme sexual selection.
TL;DR: This paper provides an overview of results from a long-term (38 generations) selection experiment that developed lines from individual phenotypic selection for high or low body weight at 8 wk of age.
TL;DR: It is argued in this paper that the development of outputs from physiological research for breeding requires a good understanding of the breeding program(s) being serviced and factors affecting its performance, and the dangers of a pre-defined or narrow focus in the physiological research.
TL;DR: In this paper, the use of permutation tests was extended to validation and investigation of four predictor variable selection techniques, namely MUSEUM, GOLPE, VIP and IVS-PLS.
Abstract: Regression model validation by permutation tests was explored. Especially in cases where the model significance is doubtful, a permutation test adds crucial information which can often can be decisive for the existence of the model. The background and applicability of the test procedure are described. As an example, the use of permutation tests was extended to validation and investigation of four predictor variable selection techniques, namely MUSEUM, GOLPE, VIP and IVS-PLS. The selection methods are briefly reviewed and compared. The permutation tests were applied before, during and after variable selection. Some similarities and differences in the behaviour of the variable selection techniques were found and are commented upon. © 1996 by John Wiley & Sons, Ltd.
TL;DR: P phenotypic selection models that examine the selection of floral traits influencing several modes of selling simultaneously are presented and suggest that reproductive assurance may be more important than has been appreciated.
Abstract: The automatic selection and reproductive assurance hypotheses provide the two most general explanations for the evolution of self-pollination. Under automatic selection, self-pollination is mediated by pollen vectors and the mating system modifier experiences a transmission bias through the pollen that leads to its selection. Under reproductive assurance, self-pollination is autonomous and the mating system modifier is selected as it allows seed production when pollinators are scarce. We present phenotypic selection models that examine the selection of floral traits influencing several modes of selling simultaneously. Inferences from these models suggest that reproductive assurance may be more important than has been appreciated. Additional insight into the importance of automatic selection versus reproductive assurance may be gained by considering the distribution of neutral genetic diversity among populations within selfing species. A number of approaches are outlined for analysing patterns of neutral diversity as they pertain to the mechanism of the evolution of selfing.
TL;DR: In his Perspective, Potts argues that another force is at work: variability selection, in which certain organisms adapt to wide fluctuations in environmental conditions.
Abstract: According to natural selection theory, major adaptive change depends on consistency in the effects of Darwinian selection. In his Perspective, Potts argues that another force is at work: variability selection, in which certain organisms adapt to wide fluctuations in environmental conditions.
TL;DR: The results suggest that selection, including overdominance, has at most a weak effect at allozyme loci and cast some doubt on the widely held notion that heterozygosity and individual fitness are strongly correlated.
Abstract: Meta-analyses of published correlation coefficients between multilocus heterozygosity (MLH) and two fitness surrogates, growth rate and fluctuating asymmetry, suggested that the strength of these correlations are generally weak. A variety of plants and animals was included in the meta-analyses. A statistically homogeneous group of MLH-growth rate correlation coefficients that included both plants and animals yielded a common correlation of rz = 0.133. A common correlation of rz = -0.170 was estimated for correlations between MLH and fluctuating asymmetry in three species of salmonid fishes. These results suggest that selection, including overdominance, has at most a weak effect at allozyme loci and cast some doubt on the widely held notion that heterozygosity and individual fitness are strongly correlated.
TL;DR: This study proposes a relatively simple computational model of organizational evolution based on the genetic algorithm that includes three organizational processes: variation, adaptation, and adaptation.
Abstract: This study proposes a relatively simple computational model of organizational evolution based on the genetic algorithm. The model includes three organizational processes: variation, adaptation, and...
TL;DR: Experimental designs are especially useful in addressing evaluation questions about the effectiveness and impact of programs as discussed by the authors, emphasizing the use of comparative data as context for interpreting findings, experimental designs increase our confidence that observed outcomes are the result of a given program or innovation instead of a function of extraneous variables or events.
Abstract: Experimental designs are especially useful in addressing evaluation questions about the effectiveness and impact of programs. Emphasizing the use of comparative data as context for interpreting findings, experimental designs increase our confidence that observed outcomes are the result of a given program or innovation instead of a function of extraneous variables or events. For example, experimental designs help us to answer such questions as the following: Would adopting a new integrated reading program improve student performance? Is TQM having a positive impact on student achievement and faculty satisfaction? Is the parent involvement program influencing parents' engagement in and satisfaction with schools? How is the school's professional development program influencing teacher's collegiality and classroom practice?