Evolutionary computation : a unified approach / Kenneth A. De Jong.

By: De Jong, Kenneth A [author].

Material type: materialTypeLabel

BookSeries: Bradford book: Publisher: Cambridge : Massachusetts Institute of Technology, [2006]Publisher: ©2006Description: ix, 256 pages : illustrations ; 24 cm.Content type: text Media type: unmediated Carrier type: volumeISBN: 0262041944 (hardback); 9780262041942 (hardback).Subject(s): Evolutionary programming (Computer science)

| Evolutionary computation

DDC classification: 005.1

Contents:

Introduction -- A historical perspective -- Canonical evolutionary algorithms -- A unified view of simple EAs -- Evolutionary algorithms as problem solvers -- Evolutionary computation theory -- Advanced EC topics -- The road ahead.

Holdings
Item type	Current library	Call number	Copy number	Status	Barcode
General lending	MTU Bishopstown Library Lending	005.1 (Browse shelf(Opens below))	1	Available	00112846

Total holds: 0

Enhanced descriptions from Syndetics:

A clear and comprehensive introduction to the field of evolutionary computation that takes an integrated approach.

Evolutionary computation, the use of evolutionary systems as computational processes for solving complex problems, is a tool used by computer scientists and engineers who want to harness the power of evolution to build useful new artifacts, by biologists interested in developing and testing better models of natural evolutionary systems, and by artificial life scientists for designing and implementing new artificial evolutionary worlds. In this clear and comprehensive introduction to the field, Kenneth De Jong presents an integrated view of the state of the art in evolutionary computation. Although other books have described such particular areas of the field as genetic algorithms, genetic programming, evolution strategies, and evolutionary programming, Evolutionary Computation is noteworthy for considering these systems as specific instances of a more general class of evolutionary algorithms. This useful overview of a fragmented field is suitable for classroom use or as a reference for computer scientists and engineers.

Bibliography: (pages 241-252) and index.

Table of contents provided by Syndetics

1 Introduction (p. 1)
1.1 Basic Evolutionary Processes (p. 2)
1.2 EV: A Simple Evolutionary System (p. 3)
1.3 EV on a Simple Fitness Landscape (p. 6)
1.4 EV on a More Complex Fitness Landscape (p. 15)
1.5 Evolutionary Systems as Problem Solvers (p. 19)
1.6 Exercises (p. 21)
2 A Historical Perspective (p. 23)
2.1 Early Algorithmic Views (p. 23)
2.2 The Catalytic 1960s (p. 24)
2.3 The Explorative 1970s (p. 25)
2.3.1 Evolutionary Programming (p. 25)
2.3.2 Evolution Strategies (p. 25)
2.3.3 Genetic Algorithms (p. 26)
2.4 The Exploitative 1980s (p. 27)
2.4.1 Optimization Applications (p. 27)
2.4.2 Other EA Applications (p. 28)
2.4.3 Summary (p. 28)
2.5 The Unifying 1990s (p. 29)
2.6 The Twenty-first Century: Mature Expansion (p. 29)
2.7 Summary (p. 31)
3 Canonical Evolutionary Algorithms (p. 33)
3.1 Introduction (p. 33)
3.2 EV(m,n) (p. 33)
3.3 Evolutionary Programming (p. 34)
3.4 Evolution Strategies (p. 36)
3.5 Genetic Algorithms (p. 40)
3.5.1 Multi-parent Reproduction (p. 41)
3.5.2 Universal Genetic Codes (p. 43)
3.6 Summary (p. 47)
4 A Unified View of Simple EAs (p. 49)
4.1 A Common Framework (p. 49)
4.2 Population Size (p. 50)
4.2.1 Parent Population Size m (p. 50)
4.2.2 Offspring Population Size n (p. 52)
4.3 Selection (p. 54)
4.3.1 Choosing Selection Mechanisms (p. 58)
4.3.2 Survival Selection: A Special Case (p. 59)
4.3.3 Selection Summary (p. 60)
4.4 Reproductive Mechanisms (p. 61)
4.4.1 Mutation (p. 61)
4.4.2 Recombination (p. 63)
4.4.3 Crossover or Mutation? (p. 66)
4.4.4 Representation Issues (p. 67)
4.4.5 Choosing Effective Reproductive Mechanisms (p. 68)
4.5 Summary (p. 69)
5 Evolutionary Algorithms as Problem Solvers (p. 71)
5.1 Simple EAs as Parallel Adaptive Search (p. 71)
5.1.1 Representation (p. 72)
5.1.1.1 Fixed-Length Linear Objects (p. 73)
5.1.1.2 Nonlinear Objects (p. 73)
5.1.1.3 Variable-Length Objects (p. 74)
5.1.1.4 Nonlinear, Variable-Length Objects (p. 75)
5.1.2 Reproductive Operators (p. 75)
5.1.3 Objective Fitness Evaluation (p. 76)
5.1.4 Population Sizes and Dynamics (p. 77)
5.1.5 Convergence and Stopping Criteria (p. 78)
5.1.6 Returning an Answer (p. 79)
5.1.7 Summary (p. 80)
5.2 EA-based Optimization (p. 80)
5.2.1 OPT-EAs (p. 80)
5.2.1.1 Fitness Scaling (p. 81)
5.2.1.2 Convergence and Elitism (p. 82)
5.2.1.3 Summary (p. 83)
5.2.2 Parameter Optimization (p. 83)
5.2.2.1 Phenotypic Representations and Operators (p. 84)
5.2.2.2 Genotypic Representations and Operators (p. 84)
5.2.2.3 Choosing Representations and Operators (p. 85)
5.2.2.4 Real-Valued Parameter Optimization (p. 85)
5.2.2.5 Integer-Valued Parameter Optimization (p. 93)
5.2.2.6 Symbolic Parameter Optimization (p. 95)
5.2.2.7 Non-homogeneous Parameter Optimization (p. 97)
5.2.3 Constrained Optimization (p. 97)
5.2.4 Data Structure Optimization (p. 100)
5.2.4.1 Variable-Length Data Structures (p. 102)
5.2.5 Multi-objective Optimization (p. 103)
5.2.6 Summary (p. 104)
5.3 EA-Based Search (p. 105)
5.4 EA-Based Machine Learning (p. 107)
5.5 EA-Based Automated Programming (p. 109)
5.5.1 Representing Programs (p. 109)
5.5.2 Evaluating Programs (p. 110)
5.5.3 Summary (p. 112)
5.6 EA-Based Adaptation (p. 112)
5.7 Summary (p. 113)
6 Evolutionary Computation Theory (p. 115)
6.1 Introduction (p. 115)
6.2 Analyzing EA Dynamics (p. 117)
6.3 Selection-Only Models (p. 120)
6.3.1 Non-overlapping-Generation Models (p. 120)
6.3.1.1 Uniform (Neutral) Selection (p. 121)
6.3.1.2 Fitness-Biased Selection (p. 123)
6.3.1.3 Non-overlapping-Generation Models with n [not equal] m (p. 132)
6.3.2 Overlapping-Generation Models (p. 134)
6.3.2.1 Uniform (Neutral) Selection (p. 135)
6.3.2.2 Fitness-Biased Selection (p. 136)
6.3.3 Selection in Standard EAs (p. 137)
6.3.4 Reducing Selection Sampling Variance (p. 138)
6.3.5 Selection Summary (p. 140)
6.4 Reproduction-Only Models (p. 141)
6.4.1 Non-overlapping-Generation Models (p. 141)
6.4.1.1 Reproduction for Fixed-Length Discrete Linear Genomes (p. 143)
6.4.1.2 Reproduction for Other Genome Types (p. 152)
6.4.2 Overlapping-Generation Models (p. 158)
6.4.3 Reproduction Summary (p. 159)
6.5 Selection and Reproduction Interactions (p. 160)
6.5.1 Evolvability and Price's Theorem (p. 160)
6.5.2 Selection and Discrete Recombination (p. 162)
6.5.2.1 Discrete Recombination from a Schema Perspective (p. 162)
6.5.2.2 Crossover-Induced Diversity (p. 163)
6.5.2.3 Crossover-Induced Fitness Improvements (p. 166)
6.5.3 Selection and Other Recombination Operators (p. 169)
6.5.4 Selection and Mutation (p. 171)
6.5.4.1 Mutation from a Schema Perspective (p. 172)
6.5.4.2 Mutation-Induced Diversity (p. 172)
6.5.4.3 Mutation-Induced Fitness Improvements (p. 175)
6.5.5 Selection and Other Mutation Operators (p. 177)
6.5.6 Selection and Multiple Reproductive Operators (p. 177)
6.5.7 Selection, Reproduction, and Population Size (p. 180)
6.5.7.1 Non-overlapping-Generation Models (p. 181)
6.5.7.2 Overlapping-Generation Models (p. 183)
6.5.8 Summary (p. 185)
6.6 Representation (p. 185)
6.6.1 Capturing Important Application Features (p. 185)
6.6.2 Defining Effective Reproduction Operators (p. 186)
6.6.2.1 Effective Mutation Operators (p. 187)
6.6.2.2 Effective Recombination Operators (p. 187)
6.7 Landscape Analysis (p. 188)
6.8 Models of Canonical EAs (p. 189)
6.8.1 Infinite Population Models for Simple GAs (p. 189)
6.8.2 Expected Value Models of Simple GAs (p. 191)
6.8.2.1 GA Schema Theory (p. 192)
6.8.2.2 Summary (p. 199)
6.8.3 Markov Models (p. 199)
6.8.3.1 Markov Models of Finite Population EAs (p. 200)
6.8.3.2 Markov Models of Simple GAs (p. 201)
6.8.3.3 Summary (p. 203)
6.8.4 Statistical Mechanics Models (p. 203)
6.8.5 Summary (p. 205)
6.9 Application-Oriented Theories (p. 205)
6.9.1 Optimization-Oriented Theories (p. 205)
6.9.1.1 Convergence and Rates of Convergence (p. 206)
6.9.1.2 ESs and Real-Valued Parameter Optimization Problems (p. 206)
6.9.1.3 Simple EAs and Discrete Optimization Problems (p. 207)
6.9.1.4 Optimizing with Genetic Algorithms (p. 208)
6.10 Summary (p. 209)
7 Advanced EC Topics (p. 211)
7.1 Self-adapting EAs (p. 211)
7.1.1 Adaptation at EA Design Time (p. 212)
7.1.2 Adaptation over Multiple EA Runs (p. 212)
7.1.3 Adaptation during an EA Run (p. 213)
7.1.4 Summary (p. 213)
7.2 Dynamic Landscapes (p. 213)
7.2.1 Standard EAs on Dynamic Landscapes (p. 214)
7.2.2 Modified EAs for Dynamic Landscapes (p. 215)
7.2.3 Categorizing Dynamic Landscapes (p. 216)
7.2.4 The Importance of the Rate of Change (p. 216)
7.2.5 The Importance of Diversity (p. 217)
7.2.6 Summary (p. 219)
7.3 Exploiting Parallelism (p. 219)
7.3.1 Coarse-Grained Parallel EAs (p. 220)
7.3.2 Fine-Grained Models (p. 220)
7.3.3 Summary (p. 221)
7.4 Evolving Executable Objects (p. 221)
7.4.1 Representation of Behaviors (p. 221)
7.4.2 Summary (p. 223)
7.5 Multi-objective EAs (p. 223)
7.6 Hybrid EAs (p. 224)
7.7 Biologically Inspired Extensions (p. 225)
7.7.1 Non-random Mating and Speciation (p. 225)
7.7.2 Coevolutionary Systems (p. 226)
7.7.2.1 CoEC Architectures (p. 227)
7.7.2.2 CoEC Dynamics (p. 227)
7.7.3 Generative Representations and Morphogenesis (p. 228)
7.7.4 Inclusion of Lamarckian Properties (p. 229)
7.7.5 Agent-Oriented Models (p. 229)
7.7.6 Summary (p. 230)
7.8 Summary (p. 230)
8 The Road Ahead (p. 231)
8.1 Modeling General Evolutionary Systems (p. 231)
8.2 More Unification (p. 232)
8.3 Summary (p. 232)
Appendix A Source Code Overview (p. 233)
A.1 EC1: A Very Simple EC System (p. 233)
A.1.1 EC1 Code Structure (p. 234)
A.1.2 EC1 Parameters (p. 235)
A.2 EC2: A More Interesting EC System (p. 236)
A.3 EC3: A More Flexible EC System (p. 237)
A.4 EC4: An EC Research System (p. 240)
Bibliography (p. 241)
Index (p. 253)

Reviews provided by Syndetics

CHOICE Review

Evolutionary computing (EC) is a research area within computer science that applies Darwinian principles to automated problem solving. The origins of EC can be traced back to the 1930s; they led to the development of genetic programming, evolution strategies, and genetic algorithms. Although the 1990s saw a synthesis of these studies under the umbrella term of EC, the field retains its eclectic nature. Because EC is represented by a variety of scholarly approaches, it is useful to disciplines outside computer science, such as evolutionary biology and design engineering. In this work, De Jong (George Mason Univ.), an active member of the EC community and founding editor in chief of the periodical Evolutionary Computation (MIT Press), summarizes the breadth of the current EC landscape by successfully integrating EC concepts among these disparate disciplines. Though noteworthy books and papers describing the various subspecies of EC already exist, such as monographs published under Springer's "Natural Computing Series," De Jong attempts the first integrated view of the field in a single volume. In doing so he provides an essential introduction to this rapidly changing, diverse area of study. ^BSumming Up: Highly recommended. Upper-division undergraduates through scholars/professionals. S. M. Frey Indiana State University

Author notes provided by Syndetics

Kenneth A. De Jong is Professor of Computer Science, Head of the Evolutionary Computation Laboratory, and Associate Director of the Krasnow Institute at George Mason University.