Grammatical Evolution: Evolutionary Automatic Programming in an Arbitrary Language

ISBN-13: 9781402074448 / Angielski / Twarda / 2003 / 144 str.

Michael O'Neill; Conor Ryan

Opis

Grammatical Evolution: Evolutionary Automatic Programming in an Arbitrary Language provides the first comprehensive introduction to Grammatical Evolution, a novel approach to Genetic Programming that adopts principles from molecular biology in a simple and useful manner, coupled with the use of grammars to specify legal structures in a search. Grammatical Evolution's rich modularity gives a unique flexibility, making it possible to use alternative search strategies - whether evolutionary, deterministic or some other approach - and to even radically change its behavior by merely changing the grammar supplied. This approach to Genetic Programming represents a powerful new weapon in the Machine Learning toolkit that can be applied to a diverse set of problem domains.
Beginning with an overview of the necessary background material in Genetic Programming and Molecular Biology, Grammatical Evolution: Evolutionary Automatic Programming in an Arbitrary Language outlines the current state of the art in grammatical and genotype-phenotype-based approaches. Following a description of Grammatical Evolution and its application to a number of example problems, an in-depth analysis of the approach is conducted, focusing on areas such as the degenerate genetic code, wrapping, and crossover. The book continues with a description of hot topics in Grammatical Evolution and presents possible directions for future research.

Recenzje

From the reviews:

"This is the first book written on grammatical evolution, a new technique that is receiving increasing attention and use. Therefore, the book fulfills an important role ... . The book contains a good description of grammatical evolution ... . 'Grammatical Evolution' should be useful for specialists and Ph.D. students in the field of grammatical evolution and genetic programming, and people working in artificial intelligence and genetic algorithms in general. We would advise it as a good resource for university libraries." (Manuel Alfonseca and Alfonso Ortega, Genetic programming and Evolvable Machines, Vol. 5, 2004)

Spis treści

1. Introduction.- 1 Evolutionary Automatic Programming.- 2 Molecular Biology.- 3 Grammars.- 4 Outline.- 2. Survey of Evolutionary Automatic Programming.- 1 Introduction.- 2 Evolutionary Automatic Programming.- 3 Origin of the Species.- 4 Tree-based Systems.- 4.1 Genetic Programming.- 4.2 Grammar based Genetic Programming.- 4.2.1 Backus Naur Form.- 4.2.2 Cellular Encoding.- 4.2.3 Bias in GP.- 4.2.4 Genetic Programming Kernel.- 4.2.5 Combining GP and ILP.- 4.2.6 Auto-parallelisation with GP.- 5 String based GP.- 5.1 BGP.- 5.2 Machine Code Genetic Programming.- 5.3 Genetic Algorithm for Deriving Software.- 5.4 CFG/GP.- 6 Conclusions.- 3. Lessons from Molecular Biology.- 1 Introduction.- 2 Genetic Codes & Gene Expression Models.- 3 Neutral Theory of Evolution.- 4 Further Principles.- 5 Desirable Features.- 6 Conclusions.- 4. Grammatical Evolution.- 1 Introduction.- 2 Background.- 3 Grammatical Evolution.- 3.1 The Biological Approach.- 3.2 The Mapping Process.- 3.2.1 Backus Naur Form.- 3.2.2 Mapping Process Outline.- 3.3 Example Individual.- 3.4 Genetic Code Degeneracy.- 3.5 The Search Algorithm.- 4 Discussion.- 5 Conclusions.- 5. Four Examples of Grammatical Evolution.- 1 Introduction.- 2 Symbolic Regression.- 2.1 Results.- 3 Symbolic Integration.- 3.1 Results.- 4 Santa Fe Ant Trail.- 4.1 Results.- 5 Caching Algorithms.- 5.1 Results.- 6 Conclusions.- 6. Analysis of Grammatical Evolution.- 1 Introduction.- 2 Wrapping Operator.- 2.1 Results.- 2.1.1 Invalid Individuals.- 2.1.2 Cumulative Frequency of Success.- 2.1.3 Genome Lengths.- 2.2 Discussion.- 3 Degenerate Genetic Code.- 3.1 Results.- 3.1.1 Diversity Measures.- 3.2 Discussion.- 4 Removal of Wrapping and Degeneracy.- 4.1 Results.- 5 Mutation Rates.- 5.1 Results.- 6 Conclusions.- 7. Crossover in Grammatical Evolution.- 1 Introduction.- 2 Homologous Crossover.- 2.1 Experimental Approach.- 2.2 Results.- 2.3 Discussion.- 3 Headless Chicken.- 3.1 Experimental Approach.- 3.2 Results.- 3.3 Discussion.- 4 Conclusions.- 8. Extensions & Applications.- 1 Translation.- 2 Alternative Search Strategies.- 3 Grammar Defined Introns.- 4 GAUGE.- 4.1 Problems.- 4.1.1 Onemax.- 4.1.2 Results.- 4.2 Mastermind - a deceptive ordering version.- 4.2.1 Results.- 4.3 Discussion.- 4.4 Conclusions and Future Work.- 5 Chorus.- 5.1 Example Individual.- 5.2 Results.- 6 Financial Prediction.- 6.1 Trading Market Indices.- 6.1.1 Experimental Setup & Results.- 7 Adaptive Logic Programming.- 7.1 Logic Programming.- 7.2 GE and Logic Programming.- 7.2.1 Backtracking.- 7.2.2 Initialisation.- 7.3 Discussion.- 8 Sensible Initialisation.- 9 Genetic Programming.- 10 Conclusions.- 9. Conclusions & Future Work.- 1 Summary.- 2 Future Work.

O autorze

Michael O'Neill [BSc. (UCD), PhD (UL)] is a lecturer in the Department of Computer Science and Information Systems at the University of Limerick. He has over 70 publications on biologically inspired algorithms (BIAs). He coauthored the Springer title "Grammatical Evolution -- Evolutionary Automatic Programming in an Arbitrary Language", Genetic Programming Series, 2003, 160 pp., ISBN 1-4020-7444-1. He is one of the two original developers of the Grammatical Evolution algorithm, research that spawned an annual invited tutorial at the largest evolutionary computation conference and an international workshop, and is also on a number of relevant organising committees (e.g., GECCO 2005). Michael is a regular reviewer for the leading evolutionary computation (EC) journals, namely IEEE Trans. on Evolutionary Computation, MIT Press's Evolutionary Computation, and Springer's Genetic Programming and Evolvable Hardware journal.