The talks given at the Eighth Workshop on Genetic Programming Theory and Practice (GPTP 2010), held at the Center for the Study of Complex Systems at the University of Michigan, are the basis for this book.
As stated in the introduction, the emphasis of the talks (and, consequently, the papers collected in the book) is on solving techniques for important classes of problems based on, but not restricted to, the genetic programming (GP) paradigm. In fact, the attention of the 2010 talks is mainly on systems where GP algorithms play a key role.
The book consists of 14 papers and an introduction. The applications it considers clearly demonstrate the maturity of GP techniques, and their ability to efficiently address difficult problem instances.
A first group of papers (the first, second, and sixth) focuses on the problem of the evolution of software. In particular, Paper 1 presents the Fertile Darwinian Bytecode Harvester (FINCH) system for Java bytecode evolution. Paper 2 discusses the Push programming language in the context of autoconstructive evolution. Paper 6 is devoted to a survey on self-modifying Cartesian GP.
A second group of papers (the third and the tenth) introduces new problems, including the 3x3 Rubik’s cube as an example of temporal sequence learning and symbolic regression for building models of probability distribution.
Art is the common denominator of Papers 13 and 14, which discuss musical composition (and financial portfolio optimization) and evolutionary art.
A further group of papers is related to the application of GP in biology (Paper 12) and industrial modeling (Papers 9 and 11). Papers 4, 7, and 8 present the use of symbolic regression and classification.
Paper 5 covers the covariant Tarpeian method for controlling bloat in GP.
A specialized audience of experts in genetic algorithms will find state-of-the-art applications and methodologies in this book. It will also be of interest to practitioners for the large number of applications discussed, and to advanced students and researchers for the numerous opportunities for investigation and thesis topics.
