List of Book Chapters

Wolfgang Banzhaf

Older papers (from 1993 back) are represented by abstracts only and are available upon email request
We give titles and links. If you click the underlined words in a title you will see an abstract and source information of the paper. If you click the corresponding filename you will retrieve a copy.

List of Abstracts and Sources

TITLE: Self-Modifying Cartesian Genetic Programming

AUTHORS: S. Harding, J. Miller and W. Banzhaf

SOURCE: Cartesian Genetic Programming
J. Miller (Ed.), Springer, New York, CI Series, 2011, 101 - 124

ABSTRACT: Book Chapter

FILENAME: SMCGP_book2011.pdf (680 kB)




TITLE: Image Processing and Cartesian Genetic Programming

AUTHORS: L. Sekanina, S. Harding, W. Banzhaf and T. Kowaliw

SOURCE: Cartesian Genetic Programming
J. Miller (Ed.), Springer, New York, CI Series, 2011, 181 - 215

ABSTRACT: Book Chapter

FILENAME: ImageCGP_book2011.pdf (2000 kB)




TITLE: Hardware Acceleration for CGP: Graphics Processing Units

AUTHORS: S. Harding and W. Banzhaf

SOURCE: Cartesian Genetic Programming
J. Miller (Ed.), Springer, New York, CI Series, 2011, 231 - 253

ABSTRACT: Book Chapter

FILENAME: GPUCGP_book2011.pdf (776 kB)




TITLE: The Use of Evolutionary Computation in Knowledge Discovery: The Example of Intrusion Detection Systems

AUTHORS: X. Wu and W. Banzhaf

SOURCE: Knowledge Mining Using Intelligent Agents
S. Dehuri, S.-B. Cho (Eds.), Imperial College Press, London, 2010, 27 - 60

ABSTRACT: This chapter discusses the use of evolutionary computation in data min- ing and knowledge discovery by using intrusion detection systems as an example. The discussion centers around the role of EAs in achieving the two high-level primary goals of data mining: prediction and descrip- tion. In particular, classi¯cation and regression tasks for prediction, and clustering tasks for description. The use of EAs for feature selection in the pre-processing step is also discussed. Another goal of this chapter was to show how basic elements in EAs, such as representations, selec- tion schemes, evolutionary operators, and ¯tness functions have to be adapted to extract accurate and useful patterns from data in diŽerent data mining tasks.

FILENAME: ( kB)




TITLE: Interday and Intraday Stock Trading using Probabilistic Adaptive Mapping Developmental Genetic Programming and Linear Genetic Programming

AUTHORS: G. Wilson and W. Banzhaf

SOURCE: Natural Computing in Computational Finance
A. Brabazon, M. O'Neill, D. Maringer (Eds.), Springer, New York, CI Series, 2010, 191 - 212

ABSTRACT: A developmental co-evolutionary genetic programming approach (PAM DGP) is compared to a standard linear genetic programming (LGP) implementation for trading of stocks in the technology sector. Both interday and intraday data for these stocks were analyzed, where both implementations were found to be impressively robust to market fluctuations while reacting efficiently to opportunities for profit. PAM DGP proved slightly more reactive to market changes compared to LGP for intraday data, where the converse held true for interday data. Both implementations had very impressive accuracy in choosing both profitable buy trades and sells that prevented losses for both interday and intraday stock data. These successful trades occurred in the context of moderately active trading for interday prices and lower levels of trading for intraday prices.

FILENAME: compfin2010.pdf (844 kB)




TITLE: A Survey of Self Modifying Cartesian Genetic Programming

AUTHORS: S. Harding, W. Banzhaf and J. Miller

SOURCE: Genetic Programming Theory and Practice VIII
R. Riolo, T. McConaghy, E. Vladislavleva (Eds.), Springer, New York, GEC Series, 2011, 91 - 107

ABSTRACT: Self-Modifying Cartesian Genetic Programming (SMCGP) is a general purpose, graph-based, developmental form of Cartesian Genetic Programming. In addition to the usual computational functions found in CGP, SMCGP includes functions that can modify the evolved program at run time. This means that programs can be iterated to produce an infinite sequence of phenotypes from a single evolved genotype. Here, we discuss the results of using SMCGP on a variety of different problems, and see that SMCGP is able to solve tasks that require scalability and plasticity. We demonstrate how SMCGP is able to produce results that would be impossible for conventional, static Genetic Programming techniques.

FILENAME: GPTP2010.pdf (768 kB)




TITLE: Algorithmic Trading with Developmental and Linear Genetic Programming

AUTHORS: G. Wilson and W. Banzhaf

SOURCE: Genetic Programming Theory and Practice VII
R. Riolo, U.-M. O'Reilly and T. McConaghy (Eds.), Springer, New York, GEC Series, 2010, 119 - 134

ABSTRACT: A developmental co-evolutionary genetic programming approach (PAM DGP) and a standard linear genetic programming (LGP) stock trading systemare applied to a number of stocks across market sectors. Both GP techniques were found to be robust to market fluctuations and reactive to opportunities associated with stock price rise and fall, with PAMDGP generating notably greater profit in some stock trend scenarios. Both algorithms were very accurate at buying to achieve profit and selling to protect assets, while exhibiting bothmoderate trading activity and the ability to maximize or minimize investment as appropriate. The content of the trading rules produced by both algorithms are also examined in relation to stock price trend scenarios.

FILENAME: gptp2009.pdf (928 kB)




TITLE: Accelerating Genetic Programming on Graphics Processing Units

AUTHORS: W. Banzhaf, S. Harding, W.B. Langdon and G. Wilson

SOURCE: Genetic Programming Theory and Practice VI
R. Riolo, T. Soule and B. Worzel (Eds.), Springer, New York, GEC Series, 2009, 229 - 248

ABSTRACT: Graphics Processing Units (GPUs) are in the process of becoming a major source of computational power for numerical applications. Originally designed for application of time-consuming graphics operations, GPUs are stream processors that implement the SIMD paradigm. The true degree of parallelism of GPUs is often hidden from the user, making programming even more flexible and convenient. In this chapter we survey Genetic Programming methods currently ported to GPUs.

FILENAME: gptp2008.pdf (1.1 MB)




TITLE: Artificial Development

AUTHORS: S. Harding and W. Banzhaf

SOURCE: Organic Computing
R.P. Wuertz (Ed.), Springer, Berlin, Complex Systems Series, 2008, 201 - 220

ABSTRACT: There is growing interest in the use of analogies of biological development for problem solving in computer science. Nature is extremely intricate when compared to human designs, and incorporates features such as the ability to scale, adapt and self-repair that could be usefully incorporated into human-designed artifacts. In this chapter, we discuss how the metaphor of biological development can be used in artificial systems and highlight some of the challenges of this emerging field.

FILENAME: evodevbookchapter.pdf (1.1 MB)




TITLE: Genetic Programming of an Algorithmic Chemistry

AUTHORS: W. Banzhaf and C. Lasarczyk

SOURCE: Genetic Programming - Theory and Applications
U.M. O'Reilly, T. Yu, R. Riolo, and B. Worzel (Eds.), Kluwer Academic, Norwell, MA, 2004, 175 - 190

ABSTRACT: We introduce a new method of execution for GP-evolved programs consisting of register machine instructions. It is shown that this method can be considered as an artificial chemistry. It lends itself well to distributed and parallel computing schemes in which synchronization and coordination are not an issue.

FILENAME: algochem.pdf (604 kB)

TITLE: On Evolutionary Design, Embodiment and Artificial Regulatory Networks

AUTHORS: W. Banzhaf

SOURCE: Embodied Artificial Intelligence, F. Iida, R. Pfeifer, L. Steels and Y. Kuniyoshi (Eds.), Springer, Berlin, LNAI 3139, 2004, 284 - 292

ABSTRACT: In this contribution we consider the idea that successful evolutionary design is best achieved in a networked system. We exemplify this thought by a discussion of artificial regulatory networks, a recently devised method to model natural genome-protein interactions. It is argued that emergent phenomena in nature require the existence of networks in order to become permanent.

FILENAME: embodied.pdf (565 kB)

TITLE: The Challenge of Complexity

AUTHORS: Wolfgang Banzhaf and Julian Miller

SOURCE: Frontiers in Evolutionary Computation
A. Menon (Ed.), Kluwer Academic, Boston, MA, 2004, 243 - 260

ABSTRACT: In this chapter we discuss the challenge provided by the problem of evolving large amounts of computer code via Genetic Programming. We argue that the problem is analogous to what Nature had to face when moving to multi-cellular life. We propose to look at developmental processes and there mechanisms to come up with solutions for this ''challenge of complexity'' in Genetic Programming.

FILENAME: challenge_rev.pdf (235 kB)




TITLE: Artificial Regulatory Networks and Genetic Programming

AUTHORS: Wolfgang Banzhaf

SOURCE: Genetic Programming - Theory and Applications
R. Riolo, B. Worzel (Eds.), Kluwer Academic, Boston, MA, 2003, 43 - 61

ABSTRACT: An artificial regulatory network able to reproduce a number of phenomena found in natural genetic regulatory networks (such as heterochrony, evolution, stability and variety of network behavior) is proposed. The connection to a new genetic representation for Genetic Programming is outlined.

FILENAME: toy_world3.pdf (1,542 kB)




TITLE: Evolving the Program for a Cell: From French Flags to Boolean Circuits

AUTHORS: J. Miller and Wolfgang Banzhaf

SOURCE: On Growth, Form and Computers, P. Bentley und S. Kumar (Eds.), Academic Press, New York, 2003, 278 - 302

FROM THE INTRODUCION ...:
The development of an entire organism from a single cell is one of the most profound and awe inspiring phenomena in the whole of the natural world. The complexity of living systems itself dwarfs anything that man has produced. This is all the more the case for the processes that lead to these intricate systems. In each phase of the development of a multi-cellular being, this living system has to survive, whether stand-alone or supported by various structures and processes provided by other living systems. Organisms construct themselves, out of humble single-celled beginnings, riding waves of interaction between the information residing in their genomes - inherited from the evolutionary past of their species via their progenitors - and the resources of their environment.

FILENAME: chapter_finalrevision.pdf (n kB)

TITLE: Optical Implementation of a Competitive Network

AUTHORS: W. Banzhaf, E. Lange, M. Oita, J. Ohta, K. Kyuma and T. Nakayama

SOURCE: Frontier Decision Support Concepts, Wiley Series in Sixth Generation Computing Technologies Wiley, New York, 1994, pp. 357 -- 390

EDITORS: V.L. Plantamura, B. Soucek, G. Visaggio

ABSTRACT:

FILENAME:

TITLE: Self-organizing Algorithms derived from RNA interactions

AUTHORS: W. Banzhaf

SOURCE: Evolution and Biocomputation, Lecture Notes in Computer Science, LNCS, Vol. 899, Springer, Berlin, 1995, pp. 69 -- 102

EDITORS: W. Banzhaf and F. Eeckman

ABSTRACT: We discuss algorithms based on the RNA interaction found in Nature. Molecular biology has reveiled that strands of RNA, besides being autocatalytic, can interact with each other. They play a double role of being information carriers and enzymes. The first role is realized by the 1-dimensional sequence of nucleotides on a strand of RNA, the second by the 3-dimensional form strands can assume under appropriate temperature and solvent conditions. We use this basic idea of having two alternative forms of the same sequence to propose a new Artificial Life algorithm. After a general introduction to the area we report our findings in a specific application studied recently: an algorithm which allows sequences of binary numbers to interact. We introduce folding methods to achieve 2-dimensional alternative forms of the sequences. Interactions between 1- and 2-dimensional forms of binary sequences generate new sequences, which compete with the original ones due to selection pressure. Starting from random sequences, replicating and self-replicating sequences are generated in considerable numbers. We follow the evolution of a number of sample simulations and analyse the resulting self-organising system.

FILENAME:




TITLE: Efficient Evolution of Machine Code for CISC Architectures using Blocks and Homologous Crossover

AUTHORS: P. Nordin, W. Banzhaf and F. Francone

SOURCE: Advances in Genetic Programming III, MIT Press, Cambridge, MA, 1999, pp. 275 -- 299

EDITORS: L. Spector, W. Langdon, U. O'Reilly and P. Angeline

ABSTRACT: This chapter describes recent advances in genetic programming of machine code. Evolutionary program induction using binary machine code is the fastest known Genetic Programming method. It is, in addition, the most well studied Genetic Programming system that uses a linear genome. Evolutionary program induction using binary machine code was originally referred to as {\em Compiling Genetic Programming System} (CGPS). For clarity, the name was changed in early 1998 to {\em Automatic Induction of Machine Code---Genetic Programming} (AIM-GP). AIM-GP stores evolved programs as linear strings of native binary machine code, which are directly executed by the processor. The absence of an interpreter and complex memory handling increases the speed of AIM-GP by about two orders of magnitude. AIM-GP has so far been applied to processors with a fixed instruction length (RISC) using integer and floating-point arithmetic. We also describe several recent advances in the AIM-GP technology. Such advances include enabling the induction of code for CISC processors such as the INTEL x86 as well as JAVA and many embedded processors. The new techniques also make AIM-GP more portable in general and simplify the adaptation to any processor architecture. Other additions include the use of floating point instructions, control flow instructions, ADFs and new genetic operators e.g. aligned homologous crossover. This chapter also discusses the benefits and drawbacks of register machine GP versus tree-based GP. This chapter is directed towards the practitioner, who wants to extend AIM-GP to new architectures and application domains.

FILENAME: aigp3_v6.pdf (169 kB)




TITLE: CAD Surface Reconstruction from Digitized 3D Point Data with Genetic Programming

AUTHORS: R. Keller, W. Banzhaf, J. Mehnen, and K. Weinert

SOURCE: Advances in Genetic Programming III, MIT Press, Cambridge, MA, 1999, pp. 41 -- 65

EDITORS: L. Spector, W. Langdon, U. O'Reilly and P. Angeline

ABSTRACT: Surface reconstruction is a hard key problem in the industrial core domain of computer-aided design (CAD) applications. A workpiece must be represented in some standard CAD object description format such that its representation can be efficiently used in a CAD process like redesign. To that end, a digitizing process represents the object surface as a weakly-structured discrete and digitized set of 3D points. Surface reconstruction attempts to transform this representation into an efficient CAD representation. Certain classic approaches produce inefficient reconstructions of surface areas that do not correspond to construction logic. Here, a new reconstruction principle along with empiric results is presented which yields logical and efficient representations. This principle is implemented as a Genetic-Programming/Evolution-Strategy-based software system.

FILENAME: Surreal.pdf (248 kB)




TITLE: Towards a metabolic robot control system

AUTHORS: Jens Ziegler, Peter Dittrich and Wolfgang Banzhaf

SOURCE: Proceedings International Workshop on Information Processing in Cells and Tissues (IPCAT'97) Sheffield, UK, September 1-4, 1997, Plenum Press, New York, 1998, 305 - 317 

EDITORS: W.M.L. Holcombe, R. Paton (Eds.)

ABSTRACT: The signal processing system of a cell is very robust in its dependence upon the observation of central metabolites and, on the other hand, on the hierarchical division into functional blocks. Therefore it can act as a model for the construction of robust, highly parallel and distributed control systems. We have used the metabolic paradigm as a guideline to develop a robot architecture for simple navigation tasks. Results on obstacle avoidance and light searching behavior are reported here. 

FILENAME: ipcat_final.pdf (337 kB)




TITLE: Explicitly Defined Introns and Destructive Crossover in Genetic Programming

AUTHORS:Peter Nordin, Frank Francone, Wolfgang Banzhaf

SOURCE: Advances in Genetic Programming II, P. Angeline, K. Kinnear (eds.), MIT Press, Cambridge, MA, 1996, pp. 111 --- 134

ABSTRACT: In Genetic Programming, introns play at least two substantial roles: (1) A structural protection role, allowing the population to preserve highly-fit building blocks; and (2) A global protection role, enabling an individual to protect itself almost entirely against the destructive effect of crossover. We introduce Explicitly Defined Introns into Genetic Programming. Our results suggest that the introduction of Explicitly Defined Introns can improve fitness, generalization, and CPU time. Further, Explicitly Defined Introns partially replace the role of Implicit Introns ( that is, introns that emerge from crossover and mutation without being explicitly defined as such). Finally, Explicitly Defined Introns and Implicit Introns appear, in some situations, to work in tandem to produce better training, fitness and generalization than occurs without Explicitly Defined Introns. 

FILENAME: aigp2.pdf (214 kB)




Wolfgang Banzhaf
Last updated: Jan 18, 2012