Catálogo de publicaciones - libros

An embodiment of a simple system, such as a one-dimensional map system, derived from heterarchical duality is discussed. We formalized two pairs of heterarchical layers induced by the indefiniteness of the environment and inconsistency between parts and wholeness by using category theory and applied its construction to a logistic map. From the analysis of its behavior, we universally observed 1/ spectrum for orbits and the fractal-like behavior in the dynamics of return maps. For the coupling map system, the parameter region with an on-off intermittency was clearly extended. Finally, we discuss the relationship between this model and the recent interest in morphological computations and search for a way to deal theoretically with the concept of adaptability.

This research applies the (CONE) method to the problem of evolving neural controllers in a simulated multi-robot system. The multi-robot system consists of multiple pursuer (predator) robots, and a single evader (prey) robot. The CONE method is designed to facilitate behavioral specialization in order to increase task performance in collective behavior solutions. Pursuit-Evasion is a task that benefits from behavioral specialization. The performance of prey-capture strategies derived by the CONE method, are compared to those derived by the (ESP) method. Results indicate that the CONE method effectively facilitates behavioral specialization in the team of pursuer robots. This specialization aids in the derivation of robust prey-capture strategies. Comparatively, ESP was found to be not as appropriate for facilitating behavioral specialization and effective prey-capture behaviors.

We introduce the concept of (defined by analogy to an information theoretic view of neutral evolution), and discuss how it might be used in the engineering of emergent systems. We describe preliminary results from an application to coarse graining of cellular automata.

In psychology the ‘A not B’ error, whereby infants perseverate in reaching to the location where a toy was previously hidden after it has been moved to a new location, has been the subject of fifty years research since it was first identified by Piaget [1]. This paper describes a novel implementation of the ‘A not B’ error paradigm which is used to test the notion that minimal systems evolutionary robotics modelling can be used to explore developmental process and to generate new hypotheses for test in natural experimental populations. The model demonstrates that agents controlled by plastic continuous time recurrent neural networks can perform the ‘A not B’ task and that homeostatic mediation of plasticity can produce perseverative error patterns similar to those observed in human infants. In addition, the model shows a developmental trend for the production of perseverative errors to reduce during development.

Classical evolutionary algorithms have been extremely successful at solving certain problems. But they implement a very simple model of evolutionary biology that misses out several aspects that might be exploited by more sophisticated algorithms. We have previously critiqued the traditional naïve approach to bio-inspired algorithm design, that moves straight from a simplistic description of the biology into some algorithm. Here we present a process for developing richer evolutionary algorithms abstracted from various processes of biological evolution, with a corresponding richer analogical computational structure, and indicate how that might be further abstracted.

In this paper, we explore the emergence of acoustic categories in sensor arrays. We describe a series experiments on the automatic categorization of species and individual birds using self-organizing maps. Experimental results showed that meaningful acoustic categories can arise as self-organizing processes in sensor arrays. In addition, we discuss how distributed categorization could be used for the emergence of symbolic communication in these platforms.

Bio-inspiration borrows three properties characteristic of living organisms: multicellular architecture, cellular division, and cellular differentiation. Implemented in silicon according to these properties, our self-organizing systems are able to grow, to self-replicate, and to self-repair. The growth and branching processes, performed by the so-called Tom Thumb algorithm, lead thus to the configuration and cloning mechanisms of the systems. The repair processes allow its cicatrization and regeneration mechanisms. The cellular implementation and hardware simulation of these mechanisms constitute the core of this paper.

There is a gap between neuroevolution systems employing artificial ontogeny and those being able to solve difficult control tasks. The NEON system builds on ideas of the well known NEAT neuroevolution system to make possible a stepwise transition from a direct encoding to complex genetic architectures using developmental processes.

Symbiosis, the collaboration of multiple organisms from different species, is common in nature. A related phenomenon, symbiogenesis, the creation of new species through the genetic integration of symbionts, is a powerful alternative to crossover as a variation operator in evolutionary algorithms. It has inspired several previous models that use the repeated composition of pre-adapted entities. In this paper we introduce a new algorithm utilizing this concept of symbiosis which is simpler and has a more natural interpretation when compared with previous algorithms. In addition it achieves success on a broader class of modular problems than some prior methods.

The language is a programming language with a minimal set of operations that exhibits universal computation. We present a conceptual framework, Chemical Bare Bones, to construct Bare Bones programs by programming the state transitions of a multi-functional catalytic particle. Molecular counts represent program variables, and are altered by the action of the catalytic particle. Chemical Bare Bones programs have unique properties with respect to correctness and time complexity. The Chemical Bare Bones implementation is naturally suited to parallel computation. Chemical Bare Bones programs are constructed and stochastically modeled to undertake computations such as multiplication.