This thesis investigates the evolution and use of abstract data types within Genetic Programming (GP). In genetic programming the principles of natural evolution (fitness based selection and recombination) acts on program code to automatically generate computer programs. The research in this thesis is motivated by the observation from software engineering that data abstraction (e.g. via abstract data types) is essential in programs created by human programmers. We investigate whether abstract data types can be similarly beneficial to the automatic production of programs using GP. GP can automatically "evolve" programs which solve non-trivial problems but few experiments have been reported where the evolved programs explicitly manipulate memory and yet memory is an essential component of most computer programs. So far work on evolving programs that explicitly use memory has principally used either problem specific memory models or a simple indexed memory model consisting of a single glo...

Computers that "program themselves"; science fact or fiction? Genetic Programming uses novel optimisation techniques to "evolve " simple programs; mimicking the way humans construct programs by progressively re-writing them. Trial programs are repeatedly modified in the search for "better/fitter " solutions. The underlying basis is Genetic Algorithms (GAs). Genetic Algorithms, pioneered by Holland [Hol92], Goldberg [Gol89] and others, are evolutionary search techniques inspired by natural selection (i.e survival of the fittest). GAs work with a "population " of trial solutions to a problem, frequently encoded as strings, and repeatedly select the "fitter " solutions, attempting to evolve better ones. The power of GAs is being demonstrated for an increasing range of applications; financial, imaging, VLSI circuit layout, gas pipeline control and production scheduling [Dav91]. But one of the most intriguing uses of GAs- driven by Koza [Koz92]- is automatic program generation. Genetic Programming applies GAs to a "population " of programs- typically encoded as tree-structures. Trial programs are evaluated against a "fitness function " and the best solutions selected for modification and re-evaluation. This modification-evaluation cycle is repeated

In this paper we present an approach to the interactive development of programs for image enhancement with Genetic Programming (GP) based on pseudo-colour transformations. In our approach the user drives GP by deciding which individual should be the winner in tournament selection. The presence of the user does not only allow running GP without a fitness function but it also transforms GP into a very efficient search procedure capable of producing effective solutions to reallife problems in only hundreds of evaluations. In the paper we also propose a strategy to further reduce user interaction: we record the choices made by the user in interactive runs and we later use them to build a model which can replace him/her in longer runs. Experimental results with interactive GP and with our user-modelling strategy are also reported. 1 Introduction In order to use Genetic Programming (GP) to solve a problem it is usually necessary to define a scalar fitness function or at least a criterion, i...

Abstract not found

Height elds are evolved for use in virtual environments. Interactive aesthetic selection is employed as a tness function for generating successive populations of images with a genetic algorithm. The images are represented using continuous layered pattern functions, which are based on procedural texturing techniques. The design space dened by the representation can be controllably biased toward specic formal qualities.

CONTROLLED GENETIC PROGRAMMING SEARCH FOR SOLVING DECEPTIVE PROBLEMS Korkmaz, Emin Erkan Ph.D., Department of Computer Engineering Supervisor: Assoc. Prof. Dr. Gokturk  Ucoluk March 2003, 77 pages Traditional Genetic Programming randomly combines subtrees by applying crossover. There is a growing interest in methods that can control such recombination operations. In this thesis, a new approach is presented for guiding the recombination process for Genetic Programming. The method is based on extracting the global information of the promising solutions that appear during the genetic search. The aim is to use this information to control the crossover operation afterwards. A separate control module is used to process the collected information. This module guides the search process by sending feedback to the genetic engine about the consequences of possible recombination alternatives.

In this paper we present an approach to the interactive development of programs for image enhancement with Genetic Programming (GP) based on pseudocolour transformations. In our approach the user drives GP by deciding which individual should be the winner in tournament selection. The presence of the user does not only allow running GP without a fitness function but it also transforms GP into a very efficient search procedure capable of producing effective solutions to real-life problems in only hundreds of evaluations. In the paper we also propose a strategy to further reduce user interaction: we record the choices made by the user in interactive runs and we later use them to build a model which can replace him/her in longer runs. Experimental results with interactive GP and with our user-modelling strategy are also reported.

Summary. This chapter presents an introduction to the different artistic design do-mains that make use of interactive evolutionary design approaches, the techniques they use, and many of the challenges arising. After a brief introduction to concepts and terminology common to most artificial genetic design, there is a survey of artis-tic evolutionary systems and related research for evolving images and forms. While the focus is primarily on purely aesthetic fitness landscapes, the survey also ven-tures into areas such as product design and architecture. The overview shifts from technique to application as organizational strategies, as appropriate. After briefly surveying additional information sources, the chapter concludes with a discussion of major topics of relevance to evolutionary system designers, providing context for the following chapters. It is hoped that this snapshot of the state of the field will increase exposure to projects and issues, discussion amongst participants, and ultimately the accessibility of these techniques and approaches. 1.1