Jeffrey K. Bassett

Evolving Behaviors for Cooperating Agents

A good deal of progress has been made in the past few years in the design and implementation of control programs for autonomous agents. A natural extension of this work is to consider solving difficult tasks with teams of cooperating agents. Our interest in this area is motivated in part by our involvement in a Navy-sponsored micro air vehicle (MAV) project in which the goal is to solve difficult surveillance tasks using a large team of small inexpensive autonomous air vehicles rather than a few expensive piloted vehicles. Our approach to developing control programs for these MAVs is to use evolutionary computation techniques to evolve behavioral rule sets. In this paper we describe our architecture for achieving this, and we present some of our initial results.

Looking Under the EA Hood with Price's Equation

In this paper we show how tools based on extensions of Price’s equation allow us to look inside production-level EAs to see how selection, representation, and reproductive operators interact with each other, and how these interactions affect EA performance. With such tools it is possible to understand at a deeper level how existing EAs work as well as provide support for making better design decisions involving new EC applications.

Learning Genetic Representations for Classes of Real-Valued Optimization Problems

Applying evolutionary algorithms to new problem domains is an exercise in the art of parameter tuning and design decisions. A great deal of work has investigated ways to automate the tuning of various EA parameters such as population size, mutation options, etc. However, genotype-to-phenotype mappings have typically been considered too complex to adapt automatically. We demonstrate a genetic representation learning method that uses meta-evolution to adapt a bitstring encoding for a synthetic class of real-valued optimization problems. The genetic representation we learn performs as well or better than a Gray code both on new instances of the problem class it was trained on and on problem types that it was not trained on.

Applying price's equation to survival selection

Several researchers have used Prices equation (from biology theory literature) to analyze the various components of an Evolutionary Algorithm (EA) while it is running, giving insights into the components contributions and interactions. While their results are interesting, they are also limited by the fact that Prices equation was designed to work with the averages of population fitness. The EA practitioner, on the other hand, is typically interested in the best individuals in the population, not the average.In this paper we introduce an approach to using Prices equation which instead calculates the upper tails of population distributions. By applying Prices equation to EAs that use survival selection instead of parent selection, this information is calculated automatically.

A new methodology for the GP theory toolbox

Recently Quantitative Genetics has been successfully employed to understand and improve operators in some Evolutionary Algorithms (EAs) implementations. This theory offers a phenotypic view of an algorithms behavior at a population level, and suggests new ways of quantifying and measuring concepts such as exploration and exploitation. In this paper, we extend the quantitative genetics approach for use with Genetic Programming (GP), adding it to the set of GP analysis techniques. We use it in combination with some existing diversity and bloat measurement tools to measure, analyze and predict the evolutionary behavior of several GP algorithms. GP specific benchmark problems, such as ant trail and symbolic regression, are used to provide new insight into how various evolutionary forces work in combination to affect the search process. Finally, using the tools, a multivariate phenotypic crossover operator is designed to both improve performance and control bloat on the difficult ant trail problem.

The relationship between evolvability and bloat

Bloat is a common problem with Evolutionary Algorithms (EAs) that use variable length representation. By creating unnecessarily large individuals it results in longer EA runtimes and solutions that are difficult to interpret. The causes of bloat are still uncertain, but one theory suggests that it occurs when the phenotype (e.g. behaviors) of the parents are not successfully inherited by their offspring. Noting the similarity to evolvability theory, which measures heritability of fitness, we hypothesize that reproductive operators with high evolvability will be less likely to cause bloat. We set out to design a new crossover operator for Pittsburgh approach classifier systems that has high phenotypic heritability. We saw an opportunity using the nearest neighbor representation to perform crossover cuts in phenotype space rather than on the genomes. We demonstrate that our operator tends to be less susceptible to bloat and has higher evolvability than a standard Pittsburgh approach crossover operator. Our hope is that this will lead to a general approach to reducing bloat for any representation.

Using Quantitative Genetics and Phenotypic Traits in Genetic Programming

When evolving executable objects, the primary focus is on the behavioral repertoire that objects exhibit. For an evolutionary algorithm (EA) approach to be effective, a fitness function must be devised that provides differential feedback across evolving objects and provides some sort of fitness gradient to guide an EA in useful directions. It is fairly well understood that needle-in-a-haystack fitness landscapes should be avoided (e.g., was the tasked accomplished or not), but much less well understood as to the alternatives.