Evolutionary Computation

Evolutionary computing uses genetic algorithms to solve problems through a learning process. Each cycle of the application builds on information learned in its previous run, therefore its problem-solving "evolves". In this book, the authors describe the basic principles of evolutionary computing, genetic algorithms, programming, and applications. Detailed coverage of binary and real encoding, including selection, crossover, and mutation, is included in two chapters. Discussion of evolution strategies covers strategy principles, mutations, recombination, and optimization. Applications for evolutionary computing are varied. Some of those covered in this book include: decision support, training & design of neural networks, pattern recognition, genetic programming, and cellular automata. TOC:Introduction.- Search, Optimization, Learning.- General Principles of Evolutive Algorithms.- Structure of a Genetic Algorithm.- Binary Encoding.- Usual Genetic Operators.- Real Encoding.- Optimization of Real Valued Functions.- Schemata Theory and Connected Problems.- Other Approaches concerning Genetic Algorithms.- Evolution Strategies.- Evolutionary Programming.- Applications.

Rapid advances in evolutionary computation have opened up a world of applications-a world rapidly growing and evolving. Decision making, neural networks, pattern recognition, complex optimization/search tasks, scheduling, control, automated programming, and cellular automata applications all rely on evolutionary computation.Evolutionary Computation presents the basic principles of evolutionary computing: genetic algorithms, evolution strategies, evolutionary programming, genetic programming, learning classifier systems, population models, and applications. It includes detailed coverage of binary and real encoding, including selection, crossover, and mutation, and discusses the (m+l) and (m,l) evolution strategy principles. The focus then shifts to applications: decision strategy selection, training and design of neural networks, several approaches to pattern recognition, cellular automata, applications of genetic programming, and more.

Principles of Evolutionary ComputationGenes and chromosomesEarly EC researchBasic evolutionary computation modelsOther EC approaches Structure of an evolutionary algorithmBasic evolutionary algorithmGenetic AlgorithmsProblem representation and fitness functionSearch progressBasic elements of genetic algorithmsCanonical genetic algorithmSchemata and building blocks Basic Selection Schemes in Evolutionary AlgorithmsSelection purposesFitness function Selection pressure and takeover timeProportional selectionTruncation Selection Based on Scaling and Ranking MechanismsScale transformationStatic scaling mechanismsDynamic scalingNoisy fitness functionsFitness remapping for minimization problemsRank-based selectionBinary tournamentq-tournament selectionFurther Selection Strategies Classification of selection strategiesElitist strategiesGeneration gap methodsSteady-state evolutionary algorithmsGenerational elitist strategies in GAsMichalewicz selectionBoltzmann selectionOther selection methodsGenetic drift Recombination Operators within Binary EncodingOne-point crossoverTwo-point crossoverN-point crossoverPunctuated crossoverSegmented crossoverShuffle crossoverUniform crossoverOther crossover operators and some comparisons Crossover probabilityMatingN-point crossover algorithm Selection for survival or replacementGeneral remarks about crossover within the framework of binary encodingMutation and other Search OperatorsMutation with binary encodingStrong and weak mutation operatorsNon-uniform mutationAdaptive non-uniform mutationSelf-adaptation of mutation rateCrossover versus mutationInversion operatorSelection versus variation operatorsSimple genetic algorithm revisitedSchema Theorem, Building Blocks and Related TopicsElements characterizing schemataSchema dynamicsEffect of selection on schema dynamicsEffect of recombination on schema dynamicsCombined effect of selection and recombination on schema dynamicsEffect of mutation on schema dynamicsSchema theoremBuilding blockBuilding block hypothesis and linkage problemGeneralizations of schema theoremDeceptive functionsReal-Valued EncodingReal-valued vectorsRecombination operators for real-valued encodingMutation operators for real-valued encodingHybridization, Parameter Setting and AdaptationSpecialized representation and hybridization within GAsParameter setting and adaptive GAsAdaptive GAsAdaptive Representations: Messy Genetic Algorithms, Delta Coding and Diploidic RepresentationPrinciples of messy genetic algorithmsRecombination within messy genetic operatorsMutationComputational model and results on messy GAsGeneralizations of messy GAsOther adaptive representation approachesDelta codingDiploidy and dominanceEvolution Strategies and Evolutionary ProgrammingEvolution strategies(1+1) strategyMultimembered evolution strategiesStandard mutationGenotypes including covariance matrix. Correlated mutationCauchy perturbationsEvolutionary programmingEvolutionary programming using Cauchy perturbationPopulation Models and Parallel ImplementationsNiching methodsFitness sharingCrowdingIsland and stepping stone modelsFine-grained and diffusion modelsCoevolutionBaldwin effectParallel implementation of evolutionary algorithmsGenetic ProgrammingEarly GP approachesProgram generating languageGP program structuresInitialization of tree structuresFitness calculationRecombination operatorsMutationSelectionPopulation modelsParallel implementationBasic GP algorithmLearning Classifier SystemsMichigan and Pittsburg families of learning classifier systemsMichigan classifier systemsBucket brigade algorithmPittsburgh classifier systemsFuzzy classifier systemsApplications of Evolutionary ComputationGeneral applications of evolutionary computationMain application areasOptimization and search applicationsChoosing a decision strategyNeural network training and designPattern recognition applicationsCellular automataEvolutionary algorithms versus other heuristics