Steven M. Gustafson

Diversity in genetic programming: an analysis of measures and correlation with fitness

Examines measures of diversity in genetic programming. The goal is to understand the importance of such measures and their relationship with fitness. Diversity methods and measures from the literature are surveyed and a selected set of measures are applied to common standard problem instances in an experimental study. Results show the varying definitions and behaviors of diversity and the varying correlation between diversity and fitness during different stages of the evolutionary process. Populations in the genetic programming algorithm are shown to become structurally similar while maintaining a high amount of behavioral differences. Conclusions describe what measures are likely to be important for understanding and improving the search process and why diversity might have different meaning for different problem domains.

Layered Learning in Genetic Programming for a Cooperative Robot Soccer Problem

We present an alternative to standard genetic programming (GP) that applies layered learning techniques to decompose a problem. GP is applied to subproblems sequentially, where the population in the last generation of a subproblem is used as the initial population of the next subproblem. This method is applied to evolve agents to play keep-away soccer, a subproblem of robotic soccer that requires cooperation among multiple agents in a dynnamic environment. The layered learning paradigm allows GP to evolve better solutions faster than standard GP. Results show that the layered learning GP outperforms standard GP by evolving a lower fitness faster and an overall better fitness. Results indicate a wide area of future research with layered learning in GP.

On improving genetic programming for symbolic regression

This paper reports an improvement to genetic programming (GP) search for the symbolic regression domain, based on an analysis of dissimilarity and mating. GP search is generally difficult to characterise for this domain, preventing well motivated algorithmic improvements. We first examine the ability of various solutions to contribute to the search process. Further analysis highlights the numerous solutions produced during search with no change to solution quality. A simple algorithmic enhancement is made that reduces these events and produces a statistically significant improvement in solution quality. We conclude by verifying the generalisability of these results on several other regression instances

Advanced Population Diversity Measures in Genetic Programming

This paper presents a survey and comparison of significant diversity measures in the genetic programming literature. This study builds on previous work by the authors to gain a deeper understanding of the conditions under which genetic programming evolution is successful. Three benchmark problems (Artificial Ant, Symbolic Regression and Even-5-Parity) are used to illustrate different diversity measures and to analyse their correlation with performance. Results show that measures of population diversity based on edit distances and phenotypic diversity suggest that successful evolution occurs when populations converge to a similar structure but with high fitness diversity.

Problem Difficulty and Code Growth in Genetic Programming

This paper investigates the relationship between code growth and problem difficulty in genetic programming. The symbolic regression problem domain is used to investigate this relationship using two different types of increased instance difficulty. Results are supported by a simplified model of genetic programming and show that increased difficulty induces higher selection pressure and less genetic diversity, which both contribute toward an increased rate of code growth.

The speciating island model: an alternative parallel evolutionary algorithm

This paper presents an investigation of a novel model for parallel evolutionary algorithms (EAs) based on the biological concept of species. In EA population search, new species represent solutions that could lead to good solutions but are disadvantaged due to their dissimilarity from the rest of the population. The Speciating Island Model (SIM) attempts to exploit new species when they arise by allocating them to new search processes executing on other islands (other processors). The long term goal of the SIM is to allow new species to diffuse throughout a large (conceptual) parallel computer network, where idle and unimproving processors initiate a new search process with them. In this paper, we focus on the successful identification and exploitation of new species and show that the SIM can achieve improved solution quality as compared to a canonical parallel EA.

Editorial to the first issue

The last 2 decades have seen the emergence of a large number of computational intelligence techniques derived from thenatural sciences.Newnature-inspiredproblem-solving paradigms emerged that are based on Darwinian evolution, entomology, condensed matter physics, neurobiology, immunology, etc. These paradigms, in turn, popularized search methodologies such as Genetic Algorithms, Genetic Programming, Evolution Strategies, Particle Swarm Optimization, Ant Colony Optimization, Simulated Annealing, Neural Networks, Artificial Immune Systems, etc. At the same time, other search methodologies such as Tabu-search, Scatter Search, GRASP, etc, remained “metaphor-less”. A war-of-the-method ensued and continues, often it is the case that each of these search paradigmshas a nicheflagship publication where the latest advances within the paradigm are presented. The IEEE Transactions on Evolutionary Algorithms or its twin publication on Neural Networks, the Evolutionary Computation journal, the journal of Genetic Programming and Evolvable Machines, the Swarm Intelligence Journal, etc, are examples of scientific outlets for work derived from nature-inspired principles, while the Journal of Heuristics, the Journal of Soft Computing—A Fusion of Foundations, Methodologies and Applications or the more recent International Journal ofMetaheuristics are examples of publications where the research emphasis is not necessarily on natural computation.

Crossover-Based Tree Distance in Genetic Programming

In evolutionary algorithms, distance metrics between solutions are often useful for many aspects of guiding and understanding the search process. A good distance measure should reflect the capability of the search: if two solutions are found to be close in distance, or similarity, they should also be close in the search algorithm sense, i.e., the variation operator used to traverse the search space should easily transform one of them into the other. This paper explores such a distance for genetic programming syntax trees. Distance measures are discussed, defined and empirically investigated. The value of such measures is then validated in the context of analysis (fitness-distance correlation is analyzed during population evolution) as well as guiding search (results are improved using our measure in a fitness sharing algorithm) and diversity (new insights are obtained as compared with standard measures).

Sampling of Unique Structures and Behaviours in Genetic Programming

This paper examines the sampling of unique structures and behaviours in genetic programming. A novel description of behaviour is used to better understand the solutions visited during genetic programming search. Results provide new insight about deception that can be used to improve the algorithm and demonstrate the capability of genetic programming to sample different large tree structures during the evolutionary process.

Using subtree crossover distance to investigate genetic programming dynamics

To analyse various properties of the search process of genetic programming it is useful to quantify the distance between two individuals. Using operator-based distance measures can make this analysis more accurate and reliable than using distance measures which have no relationship with the genetic operators. This paper extends a recent definition of a distance measure based on subtree crossover for genetic programming. Empirical studies are presented that show the suitability of this measure to dynamically calculate the fitness distance correlation coefficient during the evolution, to construct a fitness sharing system for genetic programming and to measure genotypic diversity in the population. These experiments confirm the accuracy of the new measure and its consistency with the subtree crossover genetic operator.