Ernesto Sanchez

Evolutionary Optimization: the GP toolkit Ernesto Sanchez 2011 edition

Marc Notes: Title from PDF t.p. (SpringerLink, viewed Apr. 27, 2011).; Includes bibliographical references (p. 177-178).; EBSCO complete collection. Review Quotes: From the reviews: The text is a handbook for uGP. It is aimed at providing the reader with all of the information required for proficient use of the tool. At the end of the book, a chapter presents a number of examples and applications. the book meets its main objective of being a reference for the uGP tool . It is effective as a starting guide for using the tool, and is also useful for discovering advanced features and exploiting the flexibility of individual representation. (Corrado Mencar, ACM Computing Reviews, March, 2012)"Table of Contents: 1. Evolutionary computation -- 1.1. Natural and artificial evolution -- 1.2. The classical paradigms -- 1.3. Genetic programming -- 2. Why yet another one evolutionary optimizer? -- 2.1. Background -- 2.2. Where to draw the lines -- 2.3. Individuals -- 2.4. Problem specification -- 2.5. Coding Techniques -- 3. The ?Gp architecture -- 3.1. Conceptual design -- 3.2. The evolutionary core -- 3.2.1. Evolutionary Operators -- 3.2.2. Population -- 3.3. The Evolutionary Cycle -- 3.3.1. Genetic operator selection -- 3.3.2. Parents selection -- 3.3.3. Offspring Generation -- 3.3.4. Individual Evaluation and Slaughtering -- 3.3.5. Termination and Aging -- 4. Advanced features -- 4.1. Self adaptation for exploration or exploitation -- 4.1.1. Self-adaptation inertia -- 4.1.2. Operator strength -- 4.3.3. Tournament size -- 4.2. Escaping local optimums -- 4.2.1. Operator activation probability -- 4.2.2. Tuning the elitism -- 4.3. Preserving diversity -- 4.3.1. Clone detection, scaling and extermination -- 4.3.2. Entropy and delta-entropy computation -- 4.3.3. Fitness holes -- 4.3.3. Population topology and multiple populations -- 4.4. Coping with the real problems -- 4.4.1. Parallel fitness evaluation -- 4.4.2. Multiple fitness -- 5. Performing an evolutionary run -- 5.1. Robot Pathfinder -- 5.2. ?Gp Settings -- 5.3. Population Settings -- 5.4. Library of Constraints -- 5.5. Launching the experiment -- 5.6. ?Gp Extractor -- 6. Command line syntax -- 6.1. Starting a run -- 6.2. Controlling messages to the user -- 6.3. Getting help and information -- 6.4. Controlling logging -- 6.5. Controlling recovery -- 6.6. Controlling evolution -- 6.7. Controlling evaluation -- 7. Syntax of the settings file -- 7.1. Controlling evolution -- 7.2. Controlling logging -- 7.3. Controlling recovery -- 8. Syntax of the population parameters file -- 8.1. Strategy parameters -- 8.1.1. Base parameters -- 8.1.2. Parameters for self adaptation -- 8.1.3. Other parameters -- 9. Syntax of the external constraints file -- 9.1. Purposes of the constraints -- 9.2. Organization of constraints and hierarchy -- 9.3. Specifying the structure of the individual -- 9.4. Specifying the contents of the individual -- 10. Writing a compliant evaluator -- 10.1. Information from ?Gp to the fitness evaluator -- 10.2. Expected fitness format -- 10.2.1. Good Examples -- 10.2.2. Bad Examples -- 11. Implementation details -- 11.1. Design principles -- 11.2. Architectural choices -- 11.2.1. The Graph library -- 11.2.2. The Evolutionary Core library -- 11.2.3. Front end -- 11.3. Code organization and class model -- 12. Examples and applications -- 12.1. Classical one-max -- 12.1.1. Fitness evaluator -- 12.1.2. Constraints -- 12.1.3. Population settings -- 12.1.4. ?Gp settings -- 12.1.5. Running -- 12.2. Values of parameters and their influence on the evolution: Arithmetic expressions / 134 -- 12.2.1. De Jong 3 -- 12.2.2. De Jong 4-Modified -- 12.2.3. Carrom -- 12.3. Complex individuals' structures and evaluation: Bit-counting in Assembly -- 12.3.1. Assembly individuals representation -- 12.3.2. Evaluator -- 12.3.3. Running -- Argument and option synopsis -- External constraints synopsis -- References. Publisher Marketing: This book describes an award-winning evolutionary algorithm that outperformed experts and conventional heuristics in solving several industrial problems. It presents a discussion of the theoretical and practical aspects that enabled GP (MicroGP) to autonomously find the optimal solution of hard problems, handling highly structured data, such as full-fledged assembly programs, with functions and interrupt handlers. For a practitioner, GP is simply a versatile optimizer to tackle most problems with limited setup effort. The book is valuable for all who require heuristic problem-solving methodologies, such as engineers dealing with verification and test of electronic circuits; or researchers working in robotics and mobile communication. Examples are provided to guide the reader through the process, from problem definition to gathering results. For an evolutionary computation researcher, GP may be regarded as a platform where new operators and strategies can be easily tested. MicroGP (the toolkit) is an active project hosted by Sourceforge: http: //ugp3.sourceforge.net/"