Jens Ziegler

Artificial chemistries—a review

This article reviews the growing body of scientific work in artificial chemistry. First, common motivations and fundamental concepts are introduced. Second, current research activities are discussed along three application dimensions: modeling, information processing, and optimization. Finally, common phenomena among the different systems are summarized. It is argued here that artificial chemistries are the right stuff for the study of prebiotic and biochemical evolution, and they provide a productive framework for questions regarding the origin and evolution of organizations in general. Furthermore, artificial chemistries have a broad application range of practical problems, as shown in this review.

Evolving control metabolism for a robot

This article demonstrates a new method of programming artificial chemistries. It uses the emerging capabilities of the systems dynamics for information-processing purposes. By evolution of metabolisms that act as control programs for a small robot one achieves the adaptation of the internal metabolic pathways as well as the selection of the most relevant available exteroceptors. The underlying artificial chemistry evolves efficient information-processing pathways with most benefit for the desired task, robot navigation. The results show certain relations to such biological systems as motile bacteria.

Decreasing the number of evaluations in evolutionary algorithms by using a meta-model of the fitness function

In this paper a method is presented that decreases the necessary number of evaluations in Evolutionary Algorithms. A classifier with confidence information is evolved to replace time consuming evaluations during tournament selection. Experimental analysis of a mathematical example and the application of the method to the problem of evolving walking patterns for quadruped robots show the potential of the presented approach.

Advances in Artificial Life

Self-replication is a fundamental property of many interesting physical, formal and biological systems, such as crystals, waves, automata, and especially forms of natural and artificial life. Despite its importance to many phenomena, self-replication has not been consistently defined or quantified in a rigorous, universal way. In this paper we propose a universal, continuously valued property of the interaction between a system and its environment. This property represents the effect of the presence of such a system upon the future presence of similar systems. We demonstrate both analytical and computational analysis of self-replicability factors for three distinct systems involving both discrete and continuous behaviors. 1 Overview and History Self-replication is a fundamental property of many interesting physical, formal, and biological systems, such as crystals, waves, automata, and especially forms of natural and artificial life [1]. Despite its importance to many phenomena, self-replication has not been consistently defined or quantified in a rigorous, universal way. In this paper we propose a universal, continuous valued property of the interaction between a system and its environment. This property represents the effect of the presence of such a system upon the future presence of similar systems. Subsequently, we demonstrate both analytical and computational analysis of self-replicability factors for three distinct systems involving both discrete and continuous behaviors. Two prominent issues arise in examining how self-replication has been handled when trying to extend the concept universally: how to deal with non-ideal systems and how to address so-called ‘trivial’ cases [2,3]. Moore [4] requires that in order for a configuration to be considered self-reproducing it must be capable of causing arbitrarily many offspring; this requirement extends poorly to finite environments. Lohn and Reggia [5] put forward several cellular-automata (CA) -specific definitions, and result in a binary criterion. A second issue that arose in the consideration of self-replicating automata was that some cases seemed too trivial for consideration, such as an ‘all-on’ CA, resulting in a requirement for Turing-universality [6]. 2 B. Adams and H. Lipson The definition for self-replicability we propose here is motivated in part by (a) A desire to do more than look at self-replication as a binary property applicable only to certain automata, and, (b) The goal of encapsulating a general concept in a means not reliant upon (but compatible with) ideal conditions. We wish to do this by putting self-replication on a scale that is algorithmically calculable, quantifiable, and continuous. Such a scale would allow for comparisons, both between the same system in different environments, determining ideal environments for a system’s replication, as well as between different systems in the same environment, if optimizing replicability in a given environment is desired. Rather than viewing self-replicability as a property purely of the system in question, we view it as a property of the interaction between a system and its environment. Self-Replication, as we present it, is a property embedded and based upon information, rather than a specific material framework. We construct replicability as a property relative to two different environments, which indicates the degree to which one environment yields a higher presence of the system over time. Self-replicability, then, is a comparison between an environment lacking the system and an environment in which the system is present. We will first introduce a number of definitions, and then give examples of replicability of three types of systems.

Towards a metabolic robot control system

Bacteria must be able to detect rapid changes in their environment and to adapt their metabolism to external fluctuations. They monitor their surroundings with membrane-bound and intra-cellular sensors. The regulatory mechanisms of bacteria can be seen as a model for the design of robust control systems based on an artificial chemistry. The capability to process information which is needed to keep autonomous agents surviving in unknown environments is discussed.