Accepted Papers

MODELLING ADAPTIVE AND COGNITIVE SYSTEMS (ADAPCOG 2008)

Accepted Papers

Emergent Instinctive Behaviour in Evolved Neuro-inspired Agents

Gul Muhammad Khan, Julian F.Miller, and David M.Halliday

Electronics Department, University of York, York, UK

Abstract. We investigate the emergence of intelligent behaviour of an agent in the classic AI learning environment known as Wumpus World. The agent is given a ’brain’ that is a new type of developmental neuro-inspired computational network. The neurons are defined by a genotype consisting of seven computational functions that represent electrical and developmental processes inside neuron. The computational functions are evolved using a form of Genetic Programming known as Cartesian Genetic Programming. The ’brain’ that occurs by running the genetic programs has a highly dynamic morphology in which neurons grow, and die, and neurite branches together with synaptic connections form and change in response to situations encountered in the external environment. We present results and analyse characteristics of the model and find that the agents exhibit ’instinctive’ behaviour and develop memory.

First steps toward a cognitive architecture based on adaptive automata

Joao Eduardo Kogler Junior and Reginaldo Inojosa Filho

University of Sao Paulo, Escola Politecnica, Sao Paulo SP, Brazil

Abstract. We introduce the use of adaptive automata as a basis for information representation to build a cognitive architecture. We discuss some fundamental issues of the cognitive processes from the conceptual and functional points of view, with the purpose of setting an operational framework to develop methodologies for simulation and project of cognitive characters and robots.

A Memory Model for Cognitive Agents

Guilherme Bittencourt

Departamento de Automa cão e Sistemas, Universidade Federal de Santa Catarina, Florianópolis - SC - Brazil

Abstract. In this paper we present a memory model that, syntactically, consists of logical propositions but whose semantic description includes, besides the usual truth value assignments, what we call emotional flavors, that correspond to the state of the agent’s body translated into cognitive terms. The combination between logical propositions and emotional flavors allows the agent to memorize relevant propositions that can be used for reasoning. These propositions are represented in a specific format – prime implicants/implicates – which is enriched with annotations that explicitly store the internal relations among their literals. Based on this representation, the details of the proposed memory mechanism is described and some arguments to support its psychological plausibility are presented.

Interdisciplinary Engineering of Intelligent Systems. Some Methodological Issues

Gerd Doeben-Henisch, Ute Bauer-Wersing, Louwrence Erasmus, Ulrich Schrader, and Matthias Wagner

University of Applied Sciences, Frankfurt am Main, Germany

Abstract. For the engineering of systems it is necessary to test systems during validation with regard to the given requirements specifications. This paper argues for an interdisciplinary standard which should allow the qualification of the behavior of a system to be ’intelligent’ or not. The proposed standard is characterized as an open process allowing different kinds of ’intelligence profiles’. An intelligence profile is a collection of basic intelligence units of which each unit represents an operational test case with a given task and an expected outcome (final goal state.)

Artificial Life: Prospects of a Synthetic Biology

Jon Umerez

Dept. of Logic & Philosophy of Science, University of the Basque Country (UPV/EHU)

Abstract. This paper is a short critical analysis of the theoretical ambitions of Artificial Life. The discipline is described and compared with other “enlargements” of Biology. After explaining some methodological and

conceptual aspects, the main issue is considered of whether this biology of the possible life may explain or reproduce, or to what extent, the biology of real life. The conclusion, even if positive with respect its potential contribution to a better knowledge of terrestrial life, reflects some degree of skepticism regarding the most ambitious artificial aims.

An Emotional-Evolutionary Technique for Higher Level Goal Definition in a Multi-Purpose Artificial Creature

Patrícia Rocha de Toro, Ricardo Gudwin, Mauro Sérgio Miskulin

FEEC, Universidade Estadual de Campinas - UNICAMP

Abstract. Artificial Creatures are embodied autonomous agents living in a virtual world, like e.g. in a computer game or in ethological simulation studies. Usually, these artificial creatures may have general higher-level goals, like survival, killing opponents, feeding, mating, discoverying the world, etc. These general higher-level goals must be turned into lower-level goals in time and space in order for the agent to generate its final behavior within environment. This paper presents an emotional-evolutionary technique which encodes each higher-level goal into a particular emotion, and using a blend of such emotions as a fitness function in a genetic algorithm, evolves particular goals in time and space in order to drive the creature's behavior