Kenneth A. De Jong

A Cooperative Coevolutionary Approach to Function Optimization

A general model for the coevolution of cooperating species is presented. This model is instantiated and tested in the domain of function optimization, and compared with a traditional GA-based function optimizer. The results are encouraging in two respects. They suggest ways in which the performance of GA and other EA-based optimizers can be improved, and they suggest a new approach to evolving complex structures such as neural networks and rule sets.

Cooperative Coevolution: An Architecture for Evolving Coadapted Subcomponents

To successfully apply evolutionary algorithms to the solution of increasingly complex problems, we must develop effective techniques for evolving solutions in the form of interacting coadapted subcomponents. One of the major difficulties is finding computational extensions to our current evolutionary paradigms that will enable such subcomponents to emerge rather than being hand designed. In this paper, we describe an architecture for evolving such subcomponents as a collection of cooperating species. Given a simple string- matching task, we show that evolutionary pressure to increase the overall fitness of the ecosystem can provide the needed stimulus for the emergence of an appropriate number of interdependent subcomponents that cover multiple niches, evolve to an appropriate level of generality, and adapt as the number and roles of their fellow subcomponents change over time. We then explore these issues within the context of a more complicated domain through a case study involving the evolution of artificial neural networks.

Using Genetic Algorithms for Concept Learning

In this article, we explore the use of genetic algorithms (GAs) as a key element in the design and implementation of robust concept learning systems. We describe and evaluate a GA-based system called GABIL that continually learns and refines concept classification rules from its interaction with the environment. The use of GAs is motivated by recent studies showing the effects of various forms of bias built into different concept learning systems, resulting in systems that perform well on certain concept classes (generally, those well matched to the biases) and poorly on others. By incorporating a GA as the underlying adaptive search mechanism, we are able to construct a concept learning system that has a simple, unified architecture with several important features. First, the system is surprisingly robust even with minimal bias. Second, the system can be easily extended to incorporate traditional forms of bias found in other concept learning systems. Finally, the architecture of the system encourages explicit representation of such biases and, as a result, provides for an important additional feature: the ability todynamically adjust system bias. The viability of this approach is illustrated by comparing the performance of GABIL with that of four other more traditional concept learners (AQ14, C4.5, ID5R, and IACL) on a variety of target concepts. We conclude with some observations about the merits of this approach and about possible extensions.

The supraglottal articulation of prominence in English: linguistic stress as localized hyperarticulation.

The results of an articulatory investigation of the supraglottal correlates of linguistic prominence in English, and a proposal of a unified description of linguistic stress are reported. Three models of stress are evaluated: that prominence expands jaw movement, that stress expands an abstract articulatory scale involving the opening and closing of the vocal tract, and that stress involves a localized shift toward hyperarticulate speech. A corpus of x‐ray microbeam records of sensible speech is studied, within which the stress pattern is controlled and is checked by means of an intonational analysis. Jaw movement data yield similar results to earlier studies, but kinematic differences interpreted with reference to a gestural theory suggest that different subjects use different articulatory strategies to articulate stress contrasts. In addition, the jaw, lip, and tongue interact in the articulation of stress in subject dependent ways. Thus the articulation of stress should be formulated in terms of abstract articulatory goals, rather than in terms of individual articulator positioning. Finally, the data show that stress affects the articulation of nonsonority distinctions such as backness in vowels and point of articulation in consonants. A hyperarticulation model of stress is discussed in terms of these results.

An Analysis of the Interacting Roles of Population Size and Crossover in Genetic Algorithms

In this paper we present some theoretical and empirical results on the interacting roles of population size and crossover in genetic algorithms. We summarize recent theoretical results on the disruptive effect of two forms of multi-point crossover: n-point crossover and uniform crossover. We then show empirically that disruption analysis alone is not sufficient for selecting appropriate forms of crossover. However, by taking into account the interacting effects of population size and crossover, a general picture begins to emerge. The implications of these results on implementation issues and performance are discussed, and several directions for further research are suggested.

A formal analysis of the role of multi-point crossover in genetic algorithms

On the basis of early theoretical and empirical studies, genetic algorithms have typically used 1 and 2-point crossover operators as the standard mechanisms for implementing recombination. However, there have been a number of recent studies, primarily empirical in nature, which have shown the benefits of crossover operators involving a higher number of crossover points. From a traditional theoretical point of view, the most surprising of these new results relate to uniform crossover, which involves on the averageL/2 crossover points for strings of lengthL. In this paper we extend the existing theoretical results in an attempt to provide a broader explanatory and predictive theory of the role of multi-point crossover in genetic algorithms. In particular, we extend the traditional disruption analysis to include two general forms of multi-point crossover:n-point crossover and uniform crossover. We also analyze two other aspects of multi-point crossover operators, namely, their recombination potential and exploratory power. The results of this analysis provide a much clearer view of the role of multi-point crossover in genetic algorithms. The implications of these results on implementation issues and performance are discussed, and several directions for further research are suggested.

On the Choice of the Offspring Population Size in Evolutionary Algorithms

Evolutionary algorithms (EAs) generally come with a large number of parameters that have to be set before the algorithm can be used. Finding appropriate settings is a diffi- cult task. The influence of these parameters on the efficiency of the search performed by an evolutionary algorithm can be very high. But there is still a lack of theoretically justified guidelines to help the practitioner find good values for these parameters. One such parameter is the offspring population size. Using a simplified but still realistic evolutionary algorithm, a thorough analysis of the effects of the offspring population size is presented. The result is a much better understanding of the role of offspring population size in an EA and suggests a simple way to dynamically adapt this parameter when necessary.

Genetic programming needs better benchmarks

Genetic programming (GP) is not a field noted for the rigor of its benchmarking. Some of its benchmark problems are popular purely through historical contingency, and they can be criticized as too easy or as providing misleading information concerning real-world performance, but they persist largely because of inertia and the lack of good alternatives. Even where the problems themselves are impeccable, comparisons between studies are made more difficult by the lack of standardization. We argue that the definition of standard benchmarks is an essential step in the maturation of the field. We make several contributions towards this goal. We motivate the development of a benchmark suite and define its goals; we survey existing practice; we enumerate many candidate benchmarks; we report progress on reference implementations; and we set out a concrete plan for gathering feedback from the GP community that would, if adopted, lead to a standard set of benchmarks.

The influence of migration sizes and intervals on island models

A need for solving more and more complex problems drives the Evolutionary Computation community towards advanced models of Evolutionary Algorithms. One such model is the island model which, although the subject of a variety of studies, still needs additional fundamental research. In this paper we have experimentally studied the influence of various migrations sizes and intervals on island models using a set of special functions. One of the surprising observations from these experiments is that the migration interval seems to be a dominating factor, with migration size generally playing a minor role with regard to the best solution found. Additional experiments measuring genetic diversity show that too frequent migrations cause islands to dominate others and lose global diversity before they are able to exchange solutions to produce better results. Also, we observe that even small migrations already make a significant impact on the behavior of an island model and therefore the effects are comparable to those of bigger migrations. On the other hand rare migrations cause a degraded performance due to the slow convergence. Collectively, these observations provide useful guidance for island model applications.

Measurement of Population Diversity

In evolutionary algorithms (EAs), the need to efficiently measure population diversity arises in a variety of contexts, including operator adaptation, algorithm stopping and re-starting criteria, and fitness sharing. In this paper we introduce a unified measure of population diversity and define its relationship to the most common phenotypic and genotypic diversity measures. We further demonstrate that this new measure provides a new and efficient method for computing population diversity, where the cost of computation increases linearly with population size.