Ron Breukelaar

Using a genetic algorithm to evolve behavior in multi dimensional cellular automata: emergence of behavior

Cellular automata are used in many fields to generate a global behavior with local rules. Finding the rules that display a desired behavior can be a hard task especially in real world problems. This paper proposes an improved approach to generate these transition rules for multi dimensional cellular automata using a genetic algorithm, thus giving a generic way to evolve global behavior with local rules, thereby mimicking nature. Three different problems are solved using multi dimensional topologies of cellular automata to show robustness, flexibility and potential. The results suggest that using multiple dimensions makes it easier to evolve desired behavior and that combining genetic algorithms with multi dimensional cellular automata is a very powerful way to evolve very diverse behavior and has great potential for real world problems.

On interactive evolution strategies

In this paper we discuss Evolution Strategies within the context of interactive optimization. Different modes of interaction will be classified and compared. A focus will be on the suitability of the approach in cases, where the selection of individuals is done by a human user based on subjective evaluation. We compare the convergence dynamics of different approaches and discuss typical patterns of user interactions observed in empirical studies. The discussion of empirical results will be based on a survey conducted via the world wide web. A color (pattern) redesign problems from literature will be adopted and extended. The simplicity of the chosen problems allowed us to let a larger number of people participate in our study. The amount of data collected makes it possible to add statistical support to our hypothesis about the performance and behavior of different Interactive Evolution Strategies and to figure out high-performing instantiations of the approach. The behavior of the user was also compared to a deterministic selection of the best individual by the computer. This allowed us to figure out how much the convergence speed is affected by noise and to estimate the potential for accelerating the algorithm by means of advanced user interaction schemes.

Using genetic algorithms to evolve behavior in cellular automata

It is an unconventional computation approach to evolve solutions instead of calculating them. Although using evolutionary computation in computer science dates back to the 1960s, using an evolutionary approach to program other algorithms is not that well known. In this paper a genetic algorithm is used to evolve behavior in cellular automata. It shows how this approach works for different topologies and neighborhood shapes. Some different one dimensional neighborhood shapes are investigated with the genetic algorithm and yield surprisingly good results.

Self-adaptive mutation rates in genetic algorithm for inverse design of cellular automata

Self-adaptation is used a lot in Evolutionary Strategies and with great success, yet for some reason it is not the mutation adaptation of choice for Genetic Algorithms. This poster describes how a self-adaptive mutation rate was used in a Genetic Algorithms to inverse design behavioral rules for a Cellular Automata. The unique characteristics of this search space gave rise to some interesting convergence behavior that might have implications for using self-adaptive mutation rates in other Genetic Algorithm applications and might clarify why self-adaptation in Genetic Algorithms is less successful than in Evolutionary Strategies.

Evolving Transition Rules for Multi Dimensional Cellular Automata

Genetic Algorithms have been used before to evolve transition rules for one dimensional Cellular Automata (CA) to solve e.g. the majority problem and investigate communication processes within such CA [3]. In this paper, the principle is extended to multi dimensional CA, and it is demonstrated how the approach evolves transition rules for the two dimensional case with a von Neumann neighborhood. In particular, the method is applied to the binary AND and XOR problems by using the GA to optimize the corresponding rules. Moreover, it is shown how the approach can also be used for more general patterns, and therefore how it can serve as a method for calibrating and designing CA for real-world applications.

Inverse design of cellular automata by genetic algorithms: an unconventional programming paradigm

Evolving solutions rather than computing them certainly represents an unconventional programming approach. The general methodology of evolutionary computation has already been known in computer science since more than 40 years, but their utilization to program other algorithms is a more recent invention. In this paper, we outline the approach by giving an example where evolutionary algorithms serve to program cellular automata by designing rules for their iteration. Three different goals of the cellular automata designed by the evolutionary algorithm are outlined, and the evolutionary algorithm indeed discovers rules for the CA which solve these problems efficiently.