Wolfgang Banzhaf

Genetic and Evolutionary Computation - GECCO 2004

Charged particle swarm optimization (CPSO) is well suited to the dynamic search problem since inter-particle repulsion maintains population diversity and good tracking can be achieved with a simple algorithm. This work extends the application of CPSO to the dynamic problem by considering a bi-modal parabolic environment of high spatial and temporal severity. Two types of charged swarms and an adapted neutral swarm are compared for a number of different dynamic environments which include extreme ‘needle-inthe-haystack’ cases. The results suggest that charged swarms perform best in the extreme cases, but neutral swarms are better optimizers in milder environments.

Review: The use of computational intelligence in intrusion detection systems: A review

Intrusion detection based upon computational intelligence is currently attracting considerable interest from the research community. Characteristics of computational intelligence (CI) systems, such as adaptation, fault tolerance, high computational speed and error resilience in the face of noisy information, fit the requirements of building a good intrusion detection model. Here we want to provide an overview of the research progress in applying CI methods to the problem of intrusion detection. The scope of this review will encompass core methods of CI, including artificial neural networks, fuzzy systems, evolutionary computation, artificial immune systems, swarm intelligence, and soft computing. The research contributions in each field are systematically summarized and compared, allowing us to clearly define existing research challenges, and to highlight promising new research directions. The findings of this review should provide useful insights into the current IDS literature and be a good source for anyone who is interested in the application of CI approaches to IDSs or related fields.

A comparison of linear genetic programming and neural networks in medical data mining

We introduce a new form of linear genetic programming (GP). Two methods of acceleration of our GP approach are discussed: 1) an efficient algorithm that eliminates intron code and 2) a demetic approach to virtually parallelize the system on a single processor. Acceleration of runtime is especially important when operating with complex data sets, because they are occurring in real-world applications. We compare GP performance on medical classification problems from a benchmark database with results obtained by neural networks. Our results show that GP performs comparably in classification and generalization.

Artificial chemistries—a review

This article reviews the growing body of scientific work in artificial chemistry. First, common motivations and fundamental concepts are introduced. Second, current research activities are discussed along three application dimensions: modeling, information processing, and optimization. Finally, common phenomena among the different systems are summarized. It is argued here that artificial chemistries are the right stuff for the study of prebiotic and biochemical evolution, and they provide a productive framework for questions regarding the origin and evolution of organizations in general. Furthermore, artificial chemistries have a broad application range of practical problems, as shown in this review.

Genotype-Phenotype-Mapping and Neutral Variation - A Case Study in Genetic Programming

We propose the application of a genotype-phenotype mapping to the solution of constrained optimization problems. The method consists of strictly separating the search space of genotypes from the solution space of phenotypes. A mapping from genotypes into phenotypes provides for the appropriate expression of information represented by the genotypes. The mapping is constructed as to guarantee feasibility of phenotypic solutions for the problem under study. This enforcing of constraints causes multiple genotypes to result in one and the same phenotype. Neutral variants are therefore frequent and play an important role in maintaining genetic diversity. As a specific example, we discuss Binary Genetic Programming (BGP), a variant of Genetic Programming that uses binary strings as genotypes and program trees as phenotypes.

Genetic Programming 1998: Proceedings of the Third Annual Conference

Proceedings of the Annual Conferences on Genetic Programming. These proceedings present the most recent research in the field of genetic programming as well as recent research results in the fields of genetic algorithms, artificial life and evolution strategies, DNA computing, evolvable hardware, and genetic learning classifier systems.

A SIMD interpreter for genetic programming on GPU graphics cards

Mackey-Glass chaotic time series prediction and nuclear protein classification show the feasibility of evaluating genetic programming populations directly on parallel consumer gaming graphics processing units. Using a Linux KDE computer equipped with an nVidia GeForce 8800 GTX graphics processing unit card the C++ SPMD interpretter evolves programs at Giga GP operations per second (895 million GPops). We use the RapidMind general processing on GPU (GPGPU) framework to evaluate an entire population of a quarter of a million individual programs on a non-trivial problem in 4 seconds. An efficient reverse polish notation (RPN) tree based GP is given.

Fast genetic programming on GPUs

As is typical in evolutionary algorithms, fitness evaluation in GP takes the majority of the computational effort. In this paper we demonstrate the use of the Graphics Processing Unit (GPU) to accelerate the evaluation of individuals. We show that for both binary and floating point based data types, it is possible to get speed increases of several hundred times over a typical CPU implementation. This allows for evaluation of many thousands of fitness cases, and hence should enable more ambitious solutions to be evolved using GP.

The “molecular” traveling salesman

We consider a method for optimization of NP-problems motivated by natural evolution. The basic entity is a population of individuals searching in state space defined by the problem. A message exchange mechanism between individuals enables the system to proceed fast and to avoid local optima. We introduce the concept of isolated evolution to maintain a certain degree of variance in the population. The global optimum can be approached to an arbitrary degree. The method is applied to the TSP and its behavior is shown in a couple of simulations.

Some Considerations on the Reason for Bloat

A representation-less model for genetic programming is presented. The model is intended to examine the mechanisms that lead to bloat in genetic programming (GP). We discuss two hypotheses (“fitness causes bloat” and “neutral code is protective”) and perform simulations to examine the predictions deduced from these hypotheses. Our observation is that predictions from both hypotheses are realized in the simulated model.