Catálogo de publicaciones - libros

Learning is a crucial ability of intelligent agents. Rather than presenting a complete literature review, we focus in this paper on important issues surrounding the application of machine learning (ML) techniques to agents and multi-agent systems (MAS). In this discussion we move from disembodied ML over single-agent learning to full multiagent learning. In the second part of the paper we focus on the application of Inductive Logic Programming, a knowledge-based ML technique, to MAS, and present an implemented framework in which multi-agent learning experiments can be carried out.

This paper presents an introduction to reinforcement learning and relational reinforcement learning at a level to be understood by students and researchers with different backgrounds.

It gives an overview of the fundamental principles and techniques of reinforcement learning without involving a rigorous deduction of the mathematics involved through the use of an example application.

Then, relational reinforcement learning is presented as a combination of reinforcement learning with relational learning. Its advantages - such as the possibility of using structural representations, making abstraction from specific goals pursued and exploiting the results of previous learning phases - are discussed.

The contribution sketches several ways of considering systems from the position of their modularity through viewing systems without any attention focused to their modularization, then as composed from functionally specified modules, up to the post-modular systems consisting of relatively independent autonomous modules sharing a common environment and acting in it. A relatively simple, uniform and productive theoretical framework for study of the mentioned aspects of systems behavior and modularity - the framework of the theory of grammar systems - will be presented, illustrated and discussed in certain details.

The encounter between emotion research and agent-based technology is multifaceted. One the one hand, results from emotion research start to serve as role model from nature, providing inspirations for technical design criteria for individual agents at the micro level and agent groups and societies at the macro level as well as the sophisticated linkages in between them. On the other hand, they are of immediate impact in important aspects of human-agent interaction and effective social cooperation between humans and conversational interfaces. In this broad survey, we offer an appetising selection of results from different areas of emotion research.

This manuscript discusses multi-agent coordination and control using techniques inspired by the behavior of social insects. It presents a system design that enables desirable overall behavior to emerge without exposing the individual agents to the complexity and dynamics of the overall system. The research, which this paper discusses, focuses on manufacturing control. However, the approach remains applicable to the coordination and control of other types of ironware systems (e.g. traffic, supply chain...).

Virtual enterprises paradigm represents an important application field for multi-agent approaches, both in terms of modeling and infrastructure development. This article summarizes the main challenges in this field and describes several current Multi-Agent System application approaches. A particular emphasis is given to the creation and operation phases of the virtual enterprise life cycle. Several open challenges in this area are also introduced.

With the ever growing usage of the world wide IT networks, agent technologies and multiagent systems (MAS) are attracting more and more attention. (Multi-)Agent technologies aim at the design of agents that perform well in environments that are not necessarily well structured and benevolent. This article tries to give an overview over MAS applications. However, because of the lack of space and time it is not possible to make this overview comprehensive in any sense. We therefore concentrate on the application of MAS in the context of supply chain management in virtual enterprises. Additionally, we give pointers to related workan d general literature for application-oriented research of MAS. In MAS applications emergent system behaviour is one of the most interesting phenomena one can investigate. However, there is more to MAS design than the interaction between a number of agents. For an effective system behaviour we need structure and organisation. To achieve this we present the concept of holonic multiagent systems and demonstrated how it can be utilised in the selected application domain.

We present agent-based modelling and social simulation in particular application to ecosystem management. The steps in designing and building an agent-based model are discussed, as are the methodological problems typically encountered. Examples of relevant agent-based models are given. The task of integrated ecosystem management is considered and examples are given of agent-based modelling in this context. As a further illustration, consideration is given to a possible agent-based model of the Fraser River watershed in British Columbia, and to particular difficulties that it presents.

Having a robot that carries out a task for you is certainly of some help. Having a group of robots seems to be even better because in this case the task may be finished faster and more reliably. However, dealing with a group of robots can make some problems more difficult. In this paper we sketch some of the advantages and some problems that come up when dealing with groups of robots. In particular, we describe techniques as they have been developed and tested in the area of robotic soccer.

The purpose of this communication is to present an evolutionary study of cooperation between two ethnic groups. The used approach is reformulated in a form of evolutionary prisoner’s dilemma method, where a population of strategies is evolved by applying simple reproduction process with a Darwin metaphor of natural selection (a probability of selection to the reproduction is proportional to a fitness). Our computer simulations show that an application of a principle of collective quilt does not lead to an emergence of an interethnic cooperation. When an administrator is introduced, then an emergence of interethnic cooperation may be observed. Furthermore, if the ethnic groups are of very different sizes, then the principle of collective guilt may be very devastating for smaller group so that intraethnic cooperation is destroyed. The second strategy of cooperation is called the , where agents that defected within interethnic interactions are punished inside of their ethnic groups. It means, unlikely to the principle of collective guilt, there exists only one type of punishment, loosely speaking, agents are punished “personally”.