George D. Smith

A multiagent model of the UK market in electricity generation

The deregulation of electricity markets has continued apace around the globe. The best structure for deregulated markets is a subject of much debate, and the consequences of poor structural choices can be dramatic. Understanding the effect of structure on behavior is essential, but the traditional economics approaches of field studies and experimental studies are particularly hard to conduct in relation to electricity markets. This paper describes an agent based computational economics approach for studying the effect of alternative structures and mechanisms on behavior in electricity markets. Autonomous adaptive agents, using hierarchical learning classifier systems, learn through competition in a simulated model of the UK market in electricity generation. The complex agent structure was developed through a sequence of experimentation to test whether it was capable of meeting the following requirements: first, that the agents are able to learn optimal strategies when competing against nonadaptive agents; second, that the agents are able to learn strategies observable in the real world when competing against other adaptive agents; and third, that cooperation without explicit communication can evolve in certain market situations. The potential benefit of an evolutionary economics approach to market modeling is demonstrated by examining the effects of alternative payment mechanisms on the behavior of agents.

Fitness Distance Correlation and Ridge Functions

Fitness Distance Correlation has been proposed as a measure of function optimization difficulty. This paper describes a class of functions, named the Ridge Functions which, according to the measure, should be highly misleading. However, all functions tested were optimized easily by both a GA and a simple hill climbing algorithm. Scatter graph analysis of Ridge functions gave little guidance due to the large number of functions with an identical scatter graph, the majority of which are not in the class of Ridge functions and are not simple to optimize.

Real-World Applications of Evolutionary Computing

This book constitutes the refereed proceedings of six workshops on evolutionary computation held concurrently as EvoWorkshops 2000 in Edinburgh, Scotland, UK, in April 2000. nThe 37 revised papers presented were carefully reviewed and selected by the respective program committees. All in all, the book demonstrates the broad application potential of evolutionary computing in a variety of fields. In accordance with the individual workshops, the book is divided into sections on image and signal processing; systems, controls, and drives in industry; telecommunications; scheduling and timetabling; robotics; and aeronautics

Evolutionary constructive induction

Feature construction in classification is a preprocessing step in which one or more new attributes are constructed from the original attribute set, the object being to construct features that are more predictive than the original feature set. Genetic programming allows the construction of nonlinear combinations of the original features. We present a comprehensive analysis of genetic programming (GP) used for feature construction, in which four different fitness functions are used by the GP and four different classification techniques are subsequently used to build the classifier. Comparisons are made of the error rates and the size and complexity of the resulting trees. We also compare the overall performance of GP in feature construction with that of GP used directly to evolve a decision tree classifier, with the former proving to be a more effective use of the evolutionary paradigm.

The Attribute Based Hill Climber

In this paper we introduce the Attribute Based Hill Climber, a parameter-free algorithm that provides a concrete, stand-alone implementation of a little used technique from the Tabu Search literature known as “regional aspiration”. Results of applying the algorithm to two classical optimisation problems, the Travelling Salesman Problem and the Quadratic Assignment Problem, show it to be competitive with existing general purpose heuristics in these areas.

Evolutionary feature construction using information gain and gini index

Feature construction using genetic programming is carried out to study the effect on the performance of a range of classification algorithms with the inclusion of the evolved attributes. Two different fitness functions are used in the genetic program, one based on information gain and the other based on the gini index. The classification algorithms used are three classification tree algorithms, namely C5, CART, CHAID and an MLP neural network. The intention of the research is to ascertain if the decision tree classification algorithms benefit more using features constructed using a genetic programme whose fitness function incorporates the same fundamental learning mechanism as the splitting criteria of the associated decision tree.

Evolutionary Image Analysis, Signal Processing and Telecommunications

In this work we examine the applicability of an evolutionary strategy to the problem of fitting constrained second-order surfaces to both synthetic and acquired 3D data. In particular we concentrate on the Genocop III algorithm proposed by Michalewicz [8] for the optimization of constrained functions. This is a novel application of this algorithm which has demonstrably good results when applied using parametric models. Example times for convergence are given which compare the approach to standard techniques.

Game playing with autonomous adaptive agents in a simplified economic model of the UK market in electricity generation

We describe a simplified model of the UK market in electricity where autonomous adaptive agents representing electricity generation companies compete by bidding for the right to generate in a series of noncooperative games simulating scenarios seen in the real world market. We investigate the effects on agent behaviour of alterations of the settlement method and examine under what conditions cooperation to receive increased payments can emerge between the agents.

An Adaptive Agent Model for Generator Company Bidding in the UK Power Pool

This paper describes an autonomous adaptive agent model of the UK market in electricity, where the agents represent electricity generating companies. We briefly describe the UK market in electricity generation, then detail the simplifications we have made. Our current model consists of a single adaptive agent bidding against several non-adaptive agents. The adaptive agent uses a hierarchical agent structure with two Learning Classifier Systems to evolve market bidding rules to meet two objectives. We detail how the agent interacts with its environment, the particular problems this environment presents to the agent and the agent and classifier architectures we used in our experiments. We present the results and conclude that using our structure can improve performance.

The effect of evolved attributes on classification algorithms

We carry out a systematic study of the effect on the performance of a range of classification algorithms with the inclusion of attributes constructed using genetic programming. The genetic program uses information gain as the basis of its fitness. The classification algorithms used are C5, CART, CHAID and a MLP. The results show that, for the majority of the data sets used, all algorithms benefit by the inclusion of the evolved attributes. However, for one data set, whilst the performance of C5 improves, the performance of the other techniques deteriorates. Whilst this is not statistically significant, it does indicate that care must be taken when a pre-processing technique (attribute construction using GP) and the classification technique (in this case, C5) use the same fundamental technology, in this case Information Gain.