Genetic search: Analysis using fitness moments

Genetic Algorithms are efficient and robust search methods that are being employed in a plethora of applications with extremely large search spaces. The directed search mechanism employed in Genetic Algorithms performs a simultaneous and balanced, exploration of new regions in the search space and exploitation of already discovered regions.This paper introduces the notion of fitness moments for analyzing the working of Genetic Algorithms (GAs). We show that the fitness moments in any generation may be predicted from those of the initial population. Since a knowledge of the fitness moments allows us to estimate the fitness distribution of strings, this approach provides for a method of characterizing the dynamics of GAs. In particular the average fitness and fitness variance of the population in any generation may be predicted. We introduce the technique of fitness-based disruption of solutions for improving the performance of GAs. Using fitness moments, we demonstrate the advantages of using fitness-based disruption. We also present experimental results comparing the performance of a standard GA and GAs (CDGA and AGA) that incorporate the principle of fitness-based disruption. The experimental evidence clearly demonstrates the power of fitness based disruption.

On generating optimal signal probabilities for random tests: A genetic approach

Genetic Algorithms are robust search and optimization techniques. A Genetic Algorithm based approach for determining the optimal input distributions for generating random test vectors is proposed in the paper. A cost function based on the COP testability measure for determining the efficacy of the input distributions is discussed, A brief overview of Genetic Algorithms (GAs) and the specific details of our implementation are described. Experimental results based on ISCAS-85 benchmark circuits are presented. The performance pf our GA-based approach is compared with previous results. While the GA generates more efficient input distributions than the previous methods which are based on gradient descent search, the overheads of the GA in computing the input distributions are larger. To account for the relatively quick convergence of the gradient descent methods, we analyze the landscape of the COP-based cost function. We prove that the cost function is unimodal in the search space. This feature makes the cost function amenable to optimization by gradient-descent techniques as compared to random search methods such as Genetic Algorithms.

Sibling rivalry between seeds within a fruit: Some population genetic models

Competition between seeds within a fruit for parental resources is described using one-locus-two-allele models. While a �normal� allele leads to an equitable distribution of resources between seeds (a situation which also corresponds to the parental optimum), the �selfish� allele is assumed to cause the seed carrying it to usurp a higher proportion of the resources. The outcome of competition between �selfish� alleles is also assumed to lead to an asymmetric distribution of resources, the �winner� being chosen randomly. Conditions for the spread of an initially rare selfish allele and the optimal resource allocation corresponding to the evolutionarily stable strategy, derived for species with n-seeded fruits, are in accordance with expectations based on Hamilton�s inclusive fitness criteria. Competition between seeds is seen to be most intense when there are only two seeds, and decreases with increasing number of seeds, suggesting that two-seeded fruits would be rarer than one-seeded or many-seeded ones. Available data from a large number of plant species are consistent with this prediction of the model.

On an exact populationary model of genetic algorithms

Genetic algorithms (GAs) are search methods that are being employed in a multitude of applications with extremely large search spaces. Recently, there has been considerable interest among GA researchers in understanding and formalizing the working of GAs. In an earlier paper, we have introduced the notion of binomially distributed populations as the central idea behind an exact ''populationary'' model of the large-population dynamics of the GA operators for objective functions called ''functions of unitation.'' In this paper, we extend this populationary model of GA dynamics to a more general class of objective functions called functions of unitation variables. We generalize the notion of a binomially distributed population to a generalized binomially distributed population (GBDP). We show that the effects of selection, crossover, and mutation can be exactly modelled after decomposing the population into GBDPs. Based on this generalized model, we have implemented a GA simulator for functions of two unitation variables-GASIM 2, and the distributions predicted by GASIM 2 match with those obtained from actual GA runs. The generalized populationary model of GA dynamics not only presents a novel and natural way of interpreting the workings of GAs with large populations, but it also provides for an efficient implementation of the model as a GA simulator. (C) Elsevier Science Inc. 1997.

Automatic decomposition of unstructured meshes employing genetic algorithms for parallel FEM computations

Parallel execution of computational mechanics codes requires efficient mesh-partitioning techniques. These mesh-partitioning techniques divide the mesh into specified number of submeshes of approximately the same size and at the same time, minimise the interface nodes of the submeshes. This paper describes a new mesh partitioning technique, employing Genetic Algorithms. The proposed algorithm operates on the deduced graph (dual or nodal graph) of the given finite element mesh rather than directly on the mesh itself. The algorithm works by first constructing a coarse graph approximation using an automatic graph coarsening method. The coarse graph is partitioned and the results are interpolated onto the original graph to initialise an optimisation of the graph partition problem. In practice, hierarchy of (usually more than two) graphs are used to obtain the final graph partition. The proposed partitioning algorithm is applied to graphs derived from unstructured finite element meshes describing practical engineering problems and also several example graphs related to finite element meshes given in the literature. The test results indicate that the proposed GA based graph partitioning algorithm generates high quality partitions and are superior to spectral and multilevel graph partitioning algorithms.

Identification of delamination in composite beams using spectral estimation and a genetic algorithm

An efficient strategy for identification of delamination in composite beams and connected structures is presented. A spectral finite-element model consisting of a damaged spectral element is used for model-based prediction of the damaged structural response in the frequency domain. A genetic algorithm (GA) specially tailored for damage identification is derived and is integrated with finite-element code for automation. For best application of the GA, sensitivities of various objective functions with respect to delamination parameters are studied and important conclusions are presented. Model-based simulations of increasing complexity illustrate some of the attractive features of the strategy in terms of accuracy as well as computational cost. This shows the possibility of using such strategies for the development of smart structural health monitoring softwares and systems.

Machine-tool selection and operation allocation in FMS: solving a fuzzy goal-programming model using a genetic algorithm

The decision-making process for machine-tool selection and operation allocation in a flexible manufacturing system (FMS) usually involves multiple conflicting objectives. Thus, a fuzzy goal-programming model can be effectively applied to this decision problem. The paper addresses application of a fuzzy goal-programming concept to model the problem of machine-tool selection and operation allocation with explicit considerations given to objectives of minimizing the total cost of machining operation, material handling and set-up. The constraints pertaining to the capacity of machines, tool magazine and tool life are included in the model. A genetic algorithm (GA)-based approach is adopted to optimize this fuzzy goal-programming model. An illustrative example is provided and some results of computational experiments are reported.

Assessment of genetic diversity and identification of core collection in sandalwood germplasm using RAPDS

Sandalwood is an economically important aromatic tree belonging to the family Santalaceae. The trees are used mainly for their fragrant heartwood and oil that have immense potential for foreign exchange. Very little information is available on the genetic diversity in this species. Hence studies were initiated and genetic diversity estimated using RAPD markers in 51 genotypes of Santalum album procured from different geographcial regions of India and three exotic lines of S. spicatum from Australia. Eleven selected Operon primers (10mer) generated a total of 156 consistent and unambiguous amplification products ranging from 200bp to 4kb. Rare and genotype specific bands were identified which could be effectively used to distinguish the genotypes. Genetic relationships within the genotypes were evaluated by generating a dissimilarity matrix based on Ward's method (Squared Euclidean distance). The phenetic dendrogram and the Principal Component Analysis generated, separated the 51 Indian genotypes from the three Australian lines. The cluster analysis indicated that sandalwood germplasm within India constitutes a broad genetic base with values of genetic dissimilarity ranging from 15 to 91 %. A core collection of 21 selected individuals revealed the same diversity of the entire population. The results show that RAPD analysis is an efficient marker technology for estimating genetic diversity and relatedness, thereby enabling the formulation of appropriate strategies for conservation, germplasm management, and selection of diverse parents for sandalwood improvement programmes.

Population genetic structure and conservation of Asian elephants (Elephas maximus) across India

This study examines the population genetic structure of Asian elephants (Elephas maximus) across India, which harbours over half the world's population of this endangered species. Mitochondrial DNA control region sequences and allele frequencies at six nuclear DNA microsatellite markers obtained from the dung of free-ranging elephants reveal low mtDNA and typical microsatellite diversity. Both known divergent clades of mtDNA haplotypes in the Asian elephant are present in India, with southern and central India exhibiting exclusively the β clade of Fernando et al. (2000), northern India exhibiting exclusively the α clade and northeastern India exhibiting both, but predominantly the α clade. A nested clade analysis revealed isolation by distance as the principal mechanism responsible for the observed haplotype distributions within the α and β clades. Analyses of molecular variance and pairwise population FST tests based on both mitochondrial and microsatellite DNA suggest that northern-northeastern India, central India, Nilgiris (in southern India) and Anamalai-Periyar (in southern India) are four demographically autonomous population units and should be managed separately. In addition, evidence for female philopatry, male-mediated gene flow and two possible historical biogeographical barriers is described.

Recombination Drives Genetic Diversification of Streptococcus dysgalactiae Subspecies equisimilis in a Region of Streptococcal Endemicity

Infection of the skin or throat by Streptococcus dysgalactiae subspecies equisimilis (SDSE) may result in a number of human diseases. To understand mechanisms that give rise to new genetic variants in this species, we used multi-locus sequence typing (MLST) to characterise relationships in the SDSE population from India, a country where streptococcal disease is endemic. The study revealed Indian SDSE isolates have sequence types (STs) predominantly different to those reported from other regions of the world. Emm-ST combinations in India are also largely unique. Split decomposition analysis, the presence of emm-types in unrelated clonal complexes, and analysis of phylogenetic trees based on concatenated sequences all reveal an extensive history of recombination within the population. The ratio of recombination to mutation (r/m) events (11:1) and per site r/m ratio (41:1) in this population is twice as high as reported for SDSE from non-endemic regions. Recombination involving the emm-gene is also more frequent than recombination involving housekeeping genes, consistent with diversification of M proteins offering selective advantages to the pathogen. Our data demonstrate that genetic recombination in endemic regions is more frequent than non-endemic regions, and gives rise to novel local SDSE variants, some of which may have increased fitness or pathogenic potential.