zbMATH — the first resource for mathematics

Grammatical evolution. Evolutionary automatic programming in an arbitrary language. Foreword by Wolfgang Banzhaf. (English) Zbl 1027.68030
Genetic Programming Series. 4. Boston, MA: Kluwer Academic Publishers. xvi, 144 p. (2003).
The book is based on the PhD thesis of the first author. It presents Grammatical Evolution (GE) which is a branch of Genetic Programming (GP). GE differs from traditional GP in three fundamental ways: It employs linear genoms. It performs an ontogenetic mapping from the genotype to the phenotype (programs). It employs grammars to define legal structures in the phenotype space.
Genes are only used to determine which grammar rule is applied when. The genotype-phenotype distinction allows the evolution of programs in an arbitrary language. In addition to the historical context, the biological background and the thorough description of the approach there are applications to some benchmark problems (symbolic regression, symbolic integration, the Santa Fe trail, evolution of caching algorithms) and an analysis of some features (degenerate genetic code, wrapping operator, crossover genetic search operators). Finally, extensions together with applications, e.g., in the field of financial predictions and in logic programming, as well as possible directions for future research are given. The book is well legibly.

68N19 Other programming paradigms (object-oriented, sequential, concurrent, automatic, etc.)
68T05 Learning and adaptive systems in artificial intelligence
68-02 Research exposition (monographs, survey articles) pertaining to computer science