Kenneth De Jong | George Mason University (original) (raw)

Papers by Kenneth De Jong

The method of evolutionary functional testing allows it to automate testing by transforming the t... more The method of evolutionary functional testing allows it to automate testing by transforming the test case design into an optimization problem. For this aim it is necessary to define a suitable fitness function. In this paper for an autonomous parking system two different approaches for fitness functions are presented, which evaluate the quality of parking maneuver automatically. A numerical analysis shows, that the proposed area criterion supports a faster convergence of the optimization compared to the proposed distance criterion and that the proposed area criterion describes an efficient method to find functional errors in an automated way.

The previous Dagstuhl workshop on the "Theory of Evolutionary Algorithms" held in February of 200... more The previous Dagstuhl workshop on the "Theory of Evolutionary Algorithms" held in February of 2000 had a great influence on the development of this field and provided a unique opportunity for the people working in this area to interact with each other. Therefore, we had many people who were interested in the new workshop and we could not invite all who asked us. The idea was to discuss the different approaches to a theory of evolutionary algorithms. The participants were researchers with quite different scientific background. People influenced by computer science, mathematics, physics, biology, or engineering came together which led to vivid and fruitful discussions. The organizers are happy to report that 40 researchers accepted an invitation to Dagstuhl. They came from Germany (13), USA (9), England (6), Belgium (2), Austria (2), India (1), Japan (1), Mexico (1),Netherlands (1), Romania (1), Russia (1), Spain (1), and Switzerland (1). The 32 talks captured all the aspects of a theory of evolutionary algorithms, among them EA-dynamics, non-static fitness and robustness, algorithmic aspects of EAs, recombination, fitness landscapes, global performance of EAs, and schema approaches. The schedule included an evening session showing "evolution strategies in action". Besides the official schedule the participants used unscheduled time for many discussions and some informal sessions with short talks, all inspired by the special Dagstuhl atmosphere.

This chapter contains sections titled: A Common Framework, Population Size, Selection, Reproducti... more This chapter contains sections titled: A Common Framework, Population Size, Selection, Reproductive Mechanisms, Summary

IOP Publishing Ltd eBooks, Nov 26, 2004

Genetic and Evolutionary Computation Conference, Jul 13, 1999

We have developed GA-based search procedures for continuous function approximations. We h a ve de... more We have developed GA-based search procedures for continuous function approximations. We h a ve designed algorithms that perform both adaptive mesh re nement and selection of interpolations functions, both of which are necessary to achieve highly accurate function approximations while keeping the number of mesh points at a minimum. We demonstrate the feasibility of the approach b y testing it on a variety of functions.

John Wiley & Sons, Inc. eBooks, Oct 3, 2018

This chapter contains sections titled: Self-adapting EAs, Dynamic Landscapes, Exploiting Parallel... more This chapter contains sections titled: Self-adapting EAs, Dynamic Landscapes, Exploiting Parallelism, Evolving Executable Objects, Multi-objective EAs, Hybrid EAs, Biologically Inspired Extensions, Summary

Additionally, I would like to thank Thomas Jansen, as well as the entirety of Ingo Wegener's chai... more Additionally, I would like to thank Thomas Jansen, as well as the entirety of Ingo Wegener's chair at Dortmund University. Their helpful influence is clear in the first half of my dissertation, but the extent of this influence cannot possibly be fully visible. Because they deserve it and are not thanked nearly enough, I would also like to thank the staff of the Department of Computer Science at George Mason University. They are great group of people, who have always gone out of their way to help make this process a little easier for me. Likewise, I thank the people at Library and Information Services. They do a huge and irreplaceable service for those of us in deep study, and probably cannot be thanked enough. On a personal level, I thank my wife for her support. She gives me strength and self confidence. In exchange, I wash the clothes. I get the better end of that deal, I think. Finally, I want to be among the many who say that life is an adventure of learning and for every question answered, there are a dozen more unanswered questions that are uncovered. Far from being discouraging, this truism is one for which I was prepared...and one of the reasons I feel that life is worth living. This preparation, and this love of learning, I owe to my parents, who definitely cannot be thanked enough. v

CRC Press eBooks, Sep 14, 2022

Oxford University Press eBooks, Mar 10, 2005

I continue to be surprised and pleased by the dramatic growth of interest in and applications of ... more I continue to be surprised and pleased by the dramatic growth of interest in and applications of genetic algorithms (GAs) in recent years. This growth, in turn, has placed a certain amount of healthy "stress" on the field as current understanding and traditional approaches are stretched to the limit by challenging new problems and new areas of application. At the same time, other forms of evolutionary computation such as evolution strategies [50] and evolutionary programming [22], continue to mature and provide alternative views on how the process of evolution might be captured in an efficient and useful computational framework. I don't think there can be much disagreement about the fact that Holland's initial ideas for adaptive system design have played a fundamental role in the progress we have made in the past thirty years [23, 46]. So, an occasion like this is an opportunity to reflect on where the field is now, how it got there, and where it is headed. In the following sections, I will attempt to summarize the progress that has been made, and to identify critical issues that need to be addressed for continued progress in the field. The widespread availability of inexpensive digital computers in the 1960s gave rise to their increased use as a modeling and simulation tool by the scientific community. Several groups around the world including Rechenberg and Schwefel at the Technical University of Berlin [49], Fogel et al. at the University of California at Los Angeles [22], and Holland at the University of Michigan in Ann Arbor [35] were captivated by the potential of taking early simulation models of evolution a step further and harnessing these evolutionary processes in computational forms that could be used for complex computer-based problem solving. In Holland's case, the motivation was the design and implementation of robust adaptive systems, capable of dealing with an uncertain and changing environment. His view emphasized the need for systems which self-adapt over time as a function of feedback obtained from interacting with the environment in which they operate. This led to an initial family of "reproductive plans" which formed the basis for what we call "simple genetic algorithms" today, as outlined in figure 1.

Parallel Problem Solving from Nature, 1992

Genetic Algorithms (GAs) have received a great deal of attention regarding their poten tial as op... more Genetic Algorithms (GAs) have received a great deal of attention regarding their poten tial as optimization techniques for complex functions. The level of interest and success in this area has led to a number of improvements to GA-based function optimizers and a good deal of progress in characterizing the kinds of functions that are easy/hard for GAs to optim ize. With all this activity, there has been a natural tendency to equate GAs with function optimization. However, the motivating context of Holland's initial GA work was the design and implementation of robust adaptive systems. In this paper we argue that a proper under standing of GAs in this broader adaptive systems context is a necessary prerequisite for understanding their potential application to any problem domain. We then use these insights to better understand the strengths and limitations of GAs as function optimizers.

[1990] Proceedings of the 2nd International IEEE Conference on Tools for Artificial Intelligence, Dec 4, 2002

Genetic Algorithms (GAs) have traditionally been used for non-symbolic learning tasks. In this pa... more Genetic Algorithms (GAs) have traditionally been used for non-symbolic learning tasks. In this paper we consider me application of a GA to a symbolic learning task, supervised concept learning from examples. A GA concept learner (GABL) is imple mented ahat learns a concept from a set of positive and negative examples. GABL is run in a batch incremental mode to facilitate comparison with an incremental concept learner, IDSR. Preliminary results suppon ahat. despite minimal system bias, GABL is an' effective concept learner and is quite competitive with IDSR as me target concept increases in complexity.

There is an increasing level of interest in designing and implementing intelligent agents capable... more There is an increasing level of interest in designing and implementing intelligent agents capable of surviving, performing tasks, and adapting to complex and dynamic environments. These agent-oriented worlds can range from autonomous underwater surveillance vehicles to web-based softbots, and present a variety of challenges and opportunities for machine learning. The complexity and variety of these worlds suggests the need for approaches involving learning at multiple levels and integrating more than one learning methodology. A framework for describing such approaches has been developed and will be presented. Examples of the use of this framework to design agents using both symbolic and non-symbolic learning methods will be given, and will serve as the basis of a discussion of interesting open issues.

This book constitutes the refereed proceedings of the 8th workshop on the foundations of genetic ... more This book constitutes the refereed proceedings of the 8th workshop on the foundations of genetic algorithms, FOGA 2005, held in Aizu-Wakamatsu City, Japan, in January 2005. The 16 revised full papers presented provide an outstanding source of reference for the field of theoretical evolutionary computation including evolution strategies, evolutionary programming, and genetic programming, as well as the continuing growth in interactions with other fields such as mathematics, physics, and biology.

The method of evolutionary functional testing allows it to automate testing by transforming the t... more The method of evolutionary functional testing allows it to automate testing by transforming the test case design into an optimization problem. For this aim it is necessary to define a suitable fitness function. In this paper for an autonomous parking system two different approaches for fitness functions are presented, which evaluate the quality of parking maneuver automatically. A numerical analysis shows, that the proposed area criterion supports a faster convergence of the optimization compared to the proposed distance criterion and that the proposed area criterion describes an efficient method to find functional errors in an automated way.

The previous Dagstuhl workshop on the "Theory of Evolutionary Algorithms" held in February of 200... more The previous Dagstuhl workshop on the "Theory of Evolutionary Algorithms" held in February of 2000 had a great influence on the development of this field and provided a unique opportunity for the people working in this area to interact with each other. Therefore, we had many people who were interested in the new workshop and we could not invite all who asked us. The idea was to discuss the different approaches to a theory of evolutionary algorithms. The participants were researchers with quite different scientific background. People influenced by computer science, mathematics, physics, biology, or engineering came together which led to vivid and fruitful discussions. The organizers are happy to report that 40 researchers accepted an invitation to Dagstuhl. They came from Germany (13), USA (9), England (6), Belgium (2), Austria (2), India (1), Japan (1), Mexico (1),Netherlands (1), Romania (1), Russia (1), Spain (1), and Switzerland (1). The 32 talks captured all the aspects of a theory of evolutionary algorithms, among them EA-dynamics, non-static fitness and robustness, algorithmic aspects of EAs, recombination, fitness landscapes, global performance of EAs, and schema approaches. The schedule included an evening session showing "evolution strategies in action". Besides the official schedule the participants used unscheduled time for many discussions and some informal sessions with short talks, all inspired by the special Dagstuhl atmosphere.

This chapter contains sections titled: A Common Framework, Population Size, Selection, Reproducti... more This chapter contains sections titled: A Common Framework, Population Size, Selection, Reproductive Mechanisms, Summary

IOP Publishing Ltd eBooks, Nov 26, 2004

Genetic and Evolutionary Computation Conference, Jul 13, 1999

We have developed GA-based search procedures for continuous function approximations. We h a ve de... more We have developed GA-based search procedures for continuous function approximations. We h a ve designed algorithms that perform both adaptive mesh re nement and selection of interpolations functions, both of which are necessary to achieve highly accurate function approximations while keeping the number of mesh points at a minimum. We demonstrate the feasibility of the approach b y testing it on a variety of functions.

John Wiley & Sons, Inc. eBooks, Oct 3, 2018

This chapter contains sections titled: Self-adapting EAs, Dynamic Landscapes, Exploiting Parallel... more This chapter contains sections titled: Self-adapting EAs, Dynamic Landscapes, Exploiting Parallelism, Evolving Executable Objects, Multi-objective EAs, Hybrid EAs, Biologically Inspired Extensions, Summary

Additionally, I would like to thank Thomas Jansen, as well as the entirety of Ingo Wegener's chai... more Additionally, I would like to thank Thomas Jansen, as well as the entirety of Ingo Wegener's chair at Dortmund University. Their helpful influence is clear in the first half of my dissertation, but the extent of this influence cannot possibly be fully visible. Because they deserve it and are not thanked nearly enough, I would also like to thank the staff of the Department of Computer Science at George Mason University. They are great group of people, who have always gone out of their way to help make this process a little easier for me. Likewise, I thank the people at Library and Information Services. They do a huge and irreplaceable service for those of us in deep study, and probably cannot be thanked enough. On a personal level, I thank my wife for her support. She gives me strength and self confidence. In exchange, I wash the clothes. I get the better end of that deal, I think. Finally, I want to be among the many who say that life is an adventure of learning and for every question answered, there are a dozen more unanswered questions that are uncovered. Far from being discouraging, this truism is one for which I was prepared...and one of the reasons I feel that life is worth living. This preparation, and this love of learning, I owe to my parents, who definitely cannot be thanked enough. v

CRC Press eBooks, Sep 14, 2022

Oxford University Press eBooks, Mar 10, 2005

I continue to be surprised and pleased by the dramatic growth of interest in and applications of ... more I continue to be surprised and pleased by the dramatic growth of interest in and applications of genetic algorithms (GAs) in recent years. This growth, in turn, has placed a certain amount of healthy "stress" on the field as current understanding and traditional approaches are stretched to the limit by challenging new problems and new areas of application. At the same time, other forms of evolutionary computation such as evolution strategies [50] and evolutionary programming [22], continue to mature and provide alternative views on how the process of evolution might be captured in an efficient and useful computational framework. I don't think there can be much disagreement about the fact that Holland's initial ideas for adaptive system design have played a fundamental role in the progress we have made in the past thirty years [23, 46]. So, an occasion like this is an opportunity to reflect on where the field is now, how it got there, and where it is headed. In the following sections, I will attempt to summarize the progress that has been made, and to identify critical issues that need to be addressed for continued progress in the field. The widespread availability of inexpensive digital computers in the 1960s gave rise to their increased use as a modeling and simulation tool by the scientific community. Several groups around the world including Rechenberg and Schwefel at the Technical University of Berlin [49], Fogel et al. at the University of California at Los Angeles [22], and Holland at the University of Michigan in Ann Arbor [35] were captivated by the potential of taking early simulation models of evolution a step further and harnessing these evolutionary processes in computational forms that could be used for complex computer-based problem solving. In Holland's case, the motivation was the design and implementation of robust adaptive systems, capable of dealing with an uncertain and changing environment. His view emphasized the need for systems which self-adapt over time as a function of feedback obtained from interacting with the environment in which they operate. This led to an initial family of "reproductive plans" which formed the basis for what we call "simple genetic algorithms" today, as outlined in figure 1.

Parallel Problem Solving from Nature, 1992

Genetic Algorithms (GAs) have received a great deal of attention regarding their poten tial as op... more Genetic Algorithms (GAs) have received a great deal of attention regarding their poten tial as optimization techniques for complex functions. The level of interest and success in this area has led to a number of improvements to GA-based function optimizers and a good deal of progress in characterizing the kinds of functions that are easy/hard for GAs to optim ize. With all this activity, there has been a natural tendency to equate GAs with function optimization. However, the motivating context of Holland's initial GA work was the design and implementation of robust adaptive systems. In this paper we argue that a proper under standing of GAs in this broader adaptive systems context is a necessary prerequisite for understanding their potential application to any problem domain. We then use these insights to better understand the strengths and limitations of GAs as function optimizers.

[1990] Proceedings of the 2nd International IEEE Conference on Tools for Artificial Intelligence, Dec 4, 2002

Genetic Algorithms (GAs) have traditionally been used for non-symbolic learning tasks. In this pa... more Genetic Algorithms (GAs) have traditionally been used for non-symbolic learning tasks. In this paper we consider me application of a GA to a symbolic learning task, supervised concept learning from examples. A GA concept learner (GABL) is imple mented ahat learns a concept from a set of positive and negative examples. GABL is run in a batch incremental mode to facilitate comparison with an incremental concept learner, IDSR. Preliminary results suppon ahat. despite minimal system bias, GABL is an' effective concept learner and is quite competitive with IDSR as me target concept increases in complexity.

There is an increasing level of interest in designing and implementing intelligent agents capable... more There is an increasing level of interest in designing and implementing intelligent agents capable of surviving, performing tasks, and adapting to complex and dynamic environments. These agent-oriented worlds can range from autonomous underwater surveillance vehicles to web-based softbots, and present a variety of challenges and opportunities for machine learning. The complexity and variety of these worlds suggests the need for approaches involving learning at multiple levels and integrating more than one learning methodology. A framework for describing such approaches has been developed and will be presented. Examples of the use of this framework to design agents using both symbolic and non-symbolic learning methods will be given, and will serve as the basis of a discussion of interesting open issues.

This book constitutes the refereed proceedings of the 8th workshop on the foundations of genetic ... more This book constitutes the refereed proceedings of the 8th workshop on the foundations of genetic algorithms, FOGA 2005, held in Aizu-Wakamatsu City, Japan, in January 2005. The 16 revised full papers presented provide an outstanding source of reference for the field of theoretical evolutionary computation including evolution strategies, evolutionary programming, and genetic programming, as well as the continuing growth in interactions with other fields such as mathematics, physics, and biology.