- 巻冊次
-
7 ISBN 9780122081552
内容説明
Foundations of Genetic Algorithms, Volume 7 (FOGA-7) is a collection of 22 papers written by the field's leading researchers, representing the most current, state-of-the-art research both in GAs and in evolutionary computation theory in general. Much more than proceedings, this clothbound book and its companion six volumes document the bi-annual FOGA workshops since their inception in 1990. Before publication, each paper is peer reviewed, revised, and edited. Covering the variety of analysis tools and techniques that characterize the behavior of evolutionary algorithms, the FOGA series, with its brand-new volume 7, provides the single best source of reference for the theoretical work in this field.
目次
- Editorial Introduction
- Schema Analysis of OneMax Problem: Evolution Equation for First Order Schemata
- Partitioning, Epistasis, and Uncertainty
- A Schema-theory-based Extension of Geiringer's Theorem for Linear GP and Variable-length GAs under Homologous Crossover
- Bistability in a Gene Pool GA with Mutation
- The 'Crossover Landscape' and the !YHamming Landscape!| for Binary Search Spaces
- Modelling Finite Populations
- The Sensitivity of PBIL to Its Learning Rate, and How Detailed Balance Can Remove It
- Evolutionary Algorithms and the Boltzmann Distribution
- Modeling and Simulating Diploid Simple Genetic Algorithms
- On the Evolution of Phenotypic Exploration Distributions
- How many Good Programs are there? How Long are they?
- Modeling Variation in Cooperative Coevolution Using Evolutionary Game Theory
- A Mathematical Framework for the Study of Coevolution
- Guaranteeing Coevolutionary Objective Measures
- A New Framework for the Valuation of Algorithms for Black-Box Optimization
- A Study on the Performance of the (1+1)-Evolutionary Algorithm
- The Long Term Behavior of Genetic Algorithms with Stochastic Evaluation
- On the Behavior of fvfYf(1)fzfnfUEfwf{ES Optimizing Functions Disturbed by Generalized Noise
- Parameter Perturbation Mechanisms in Binary Coded GAs with Self-Adaptive Mutation
- Fitness Gains and Mutation Patterns: Deriving Mutation Rates by Exploiting Landscape Data
- Towards Qualitative Models of Interactions in Evolutionary Algorithms
- Genetic Search Reinforced by the Population Hierarchy
- 巻冊次
-
[1] ISBN 9781558601703
内容説明
Foundations of Genetic Algorithms 1991 (FOGA 1) discusses the theoretical foundations of genetic algorithms (GA) and classifier systems.
This book compiles research papers on selection and convergence, coding and representation, problem hardness, deception, classifier system design, variation and recombination, parallelization, and population divergence. Other topics include the non-uniform Walsh-schema transform; spurious correlations and premature convergence in genetic algorithms; and variable default hierarchy separation in a classifier system. The grammar-based genetic algorithm; conditions for implicit parallelism; and analysis of multi-point crossover are also elaborated. This text likewise covers the genetic algorithms for real parameter optimization and isomorphisms of genetic algorithms. This publication is a good reference for students and researchers interested in genetic algorithms.
目次
Part 1: Genetic Algorithm Hardness
The Nonuniform Walsh-Schema Transform
Epistasis Variance: A Viewpoint on GA-Hardness
Deceptiveness and Genetic Algorithm Dynamics
Part 2: Selection and Convergence
An Extension to the Theory of Convergence and a Proof of the Time Complexity of Genetic Algorithms
A Comparative Analysis of Selection Schemes Used in Genetic Algorithms
A Study of Reproduction in Generational and Steady State Genetic Algorithms
Spurious Correlations and Premature Convergence in Genetic Algorithms
Part 3: Classifier Systems
Representing Attribute-Based Concepts in a Classifier System
Quasimorphisms or Queasymorphisms? Modeling Finite Automaton Environments
Variable Default Hierarchy Separation in a Classifier System
Part 4: Coding and Representation
A Hierarchical Approach to Learning the Boolean Multiplexer Function
A Grammar-Based Genetic Algorithm
Genetic Algorithms for Real Parameter Optimization
Part 5: Framework Issues
Fundamental Principles of Deception in Genetic Search
Isomorphisms of Genetic Algorithms
Conditions for Implicit Parallelism
Part 6: Variation and Recombination
The CHC Adaptive Search Algorithm: How to Have Safe Search When Engaging in Nontraditional Genetic Recombination
Genetic Operators for Sequencing Problems
An Analysis of Multi-Point Crossover
Evolution in Time and Space-The Parallel Genetic Algorithm
Author Index
Key Word Index
- 巻冊次
-
2 ISBN 9781558602632
内容説明
Foundations of Genetic Algorithms, Volume 2 provides insight of theoretical work in genetic algorithms. This book provides a general understanding of a canonical genetic algorithm. Organized into six parts encompassing 19 chapters, this volume begins with an overview of genetic algorithms in the broader adaptive systems context. This text then reviews some results in mathematical genetics that use probability distributions to characterize the effects of recombination on multiple loci in the absence of selection. Other chapters examine the static building block hypothesis (SBBH), which is the underlying assumption used to define deception. This book discusses as well the effect of noise on the quality of convergence of genetic algorithms. The final chapter deals with the primary goal in machine learning and artificial intelligence, which is to dynamically and automatically decompose problems into simpler problems to facilitate their solution. This book is a valuable resource for theorists and genetic algorithm researchers.
目次
Introduction
Part 1: Foundation Issues Revisited
Genetic Algorithms are Not Function Optimizers
Generation Gaps Revisited
Part 2: Modeling Genetic Algorithms
Recombination Distributions for Genetic Algorithms
An Executable Model of a Simple Genetic Algorithm
Modeling Simple Genetic Algorithms
Part 3: Deception and the Building Block Hypothesis
Deception Considered Harmful
Analyzing Deception in Trap Functions
Relative Building-Block Fitness and the Building Block Hypothesis
Part 4: Convergence and Genetic Diversity
Accounting for Noise in the Sizing of Populations
Population Diversity in an Immune System Model: Implications for Genetic Search
Remapping Hyperspace During Genetic Search: Canonical Delta Folding
Part 5: Genetic Operators and Their Analysis
Real-Coded Genetic Algorithms and Interval-Schemata
Genetic Set Recombination
Crossover or Mutation?
Simulated Crossover in Genetic Algorithms
Part 6: Machine Learning
Learning Boolean Functions with Genetic Algorithms: A PAC Analysis
Is the Genetic Algorithm a Cooperative Learner?
Hierarchical Automatic Function Definition in Genetic Programming
Author Index
Key Word Index
- 巻冊次
-
3 ISBN 9781558603561
内容説明
Foundations of Genetic Algorithms, 3 focuses on the principles, methodologies, and approaches involved in the integration of genetic algorithm into mainstream mathematics, as well as genetic operators, genetic programming, and evolutionary algorithms. The selection first offers information on an experimental design perspective on genetic algorithms; schema theorem and price's theorem; and fitness variance of formae and performance prediction. Discussions focus on representation-independent recombination, representation-independent mutation and hill-climbing, recombination and the re-emergence of schemata, and Walsh transforms and deception. The publication then examines the troubling aspects of a building block hypothesis for genetic programming and order statistics for convergence velocity analysis of simplified evolutionary algorithms. The manuscript ponders on stability of vertex fixed points and applications; predictive models using fitness distributions of genetic operators; and modeling simple genetic algorithms for permutation problems. Topics include exact models for permutations, fitness distributions of genetic operators, predictive model based on linear fitness distributions, and stability in the simplex. The book also takes a look at the role of development in genetic algorithms and productive recombination and propagating and preserving schemata. The selection is a dependable source of data for mathematicians and researchers interested in genetic algorithms.
目次
?Introduction
Part 1: Schema Based Analyses
An Experimental Design Perspective on Genetic Algorithms
The Schema Theorem and Price's Theorem
Fitness Variance of Formae and Performance Prediction
The Troubling Aspects of a Building Block Hypothesis for Genetic Programming
Part 2: Convergence and Predictive Models
Order Statistics for Convergence Velocity Analysis of Simplified Evolutionary Algorithms
Stability of Vertex Fixed Points and Applications
Using Markov Chains to Analyze GAFOs
Predictive Models Using Fitness Distributions of Genetic Operators
Modeling Simple Genetic Algorithms for Permutation Problems
Population Size and Genetic Drift in Fitness Sharing
An Approach to the Study of Sensitivity for a Class of Genetic Algorithms
Part 3: Fitness Landscapes and Genetic Operators
Genetic Algorithm Difficulty and the Modality of Fitness Landscapes
Greedy Recombination and Genetic Search on the Space of Computer Programs
Productive Recombination and Propagating and Preserving Schemata
The Role of Development in Genetic Algorithms
Author Index
Key Word Index
- 巻冊次
-
5 ISBN 9781558605596
内容説明
Foundations of Genetic Algorithms, Volume 5 is the fifth in the series of books recording the prestigious Foundations of Genetic Algorithms Workshops, sponsored and organized by the International Society of Genetic Algorithms specifically to address theoretical publications on genetic algorithms and classifier systems. This volume's papers deal with GA dynamics; genetic operators (mostly in their relationship to schemata); characterization of landscapes over which an algorithm is searching; and the interaction between different parameters or strategies used for controlling the course of genetic search.
- 巻冊次
-
6 ISBN 9781558607347
内容説明
Foundations of Genetic Algorithms, Volume 6 is the latest in a series of books that records the prestigious Foundations of Genetic Algorithms Workshops, sponsored and organised by the International Society of Genetic Algorithms specifically to address theoretical publications on genetic algorithms and classifier systems.
Genetic algorithms are one of the more successful machine learning methods. Based on the metaphor of natural evolution, a genetic algorithm searches the available information in any given task and seeks the optimum solution by replacing weaker populations with stronger ones.
「Nielsen BookData」 より