ESOA 2004

 ESOA'04 Workshop The Second International Workshop on

Engineering Self-Organising Applications (ESOA'04)

to be held at the

Third International Joint Conference on Autonomous Agents and
Multi-Agent Systems ( AAMAS'04 ), July 20, 2004 - New York, USA

in conjuction with the

Tutorial on Engineering Self-Organizing Applications

Introduction
The spread of the Internet, mobile communications, and the change of traditional market models, such as in e-commerce, result in the whole information infrastructure operating as a global dynamic system. Therefore, software must adapt to personal requirements, by providing highly customised services to a huge user population. New maintenance requirements have to be met, for example software that cannot be stopped must evolve to meet changing requirements. In a large interconnected system this is a task beyond centralised management techniques. At the other end of the scale, ad-hoc sensors networks and MEMs devices, are making it possible to embed millions of smart computing agents into the environment. Here too we need the system to be able to adapt to constant failures and replacement of agents and changes in the environment, without human intervention.

A way to meet these requirements is to utilise the emergent properties of distributed interacting software -- to utilise concepts such as self-organisation, self-regulation, self-repair and self-maintenance. However, in artificial systems, environmental pressures and local interactions and control may lead to unpredicted or undesirable behaviour. Understanding how to engineer the correct self-organisation behaviour is thus an issue of major concern.

Self-organising applications (SOAs) are able to dynamically change their functionality and structure without direct user intervention to meet changes in requirements and their environment. The overall functionality delivered by SOAs typically changes progressively, mainly in a non-linear fashion, until it reaches (emerges to) a state where it satisfies the system requirements at the time and therefore it is termed self-organising or emergent behaviour.

Self-organising behaviour is often the result of the execution of a number of individual application components that locally interact with each other aiming to achieve their local goals, for example systems that are based on agents or distributed objects. The main characteristic of such systems is their ability to achieve complex collective tasks with relatively simple individual behaviours, without central or hierarchical control.

A major open issue is therefore how to engineer desirable emergent behaviour in SOAs and how to avoid undesirable one given the requirements and the application environment. To address this issue, approaches originating from diverse areas such as non-linear optimisation, knowledge-based programming and constraint problem solving are currently been explored. Furthermore, SOA engineers often take inspiration from the real world, for example from biology, chemistry, sociology and the physical world. For example, typical examples of SOAs are systems that reproduce socially-based insect behaviour, such as ants-based systems, artificial life, or robots. Although the results achieved so far are promising, further work is required until the problem is sufficiently addressed.

More specific fundamental questions that need an answer are: How do we structure the application components and their interactions, so that the self-organisation process results in the desired functionality? How do we validate that the application performs to the requirements within the range of scenarios expected during deployment? What means of influencing the dynamics of the application do we have available and how effective are they? On the one hand, multi-agent simulations and analytic modelling can be used to study emergent behaviour in real systems. On the other hand, results issued from complexity theory can be applied in engineering of both multi-agent systems and self-organising systems.

The goal of this workshop is to open a dialog among practitioners from diverse fields, including: agent based systems, information systems, distributed systems, complex systems, optimisation theory and non-linear systems, neural networks, and evolutionary computation. Last year workshop has seen great interest in applying nature-inspired models to fields as diverse as network security, manufacturing control, and electronic markets. This workshop will focus on interaction mechanisms and middleware technology specifically supporting self-organisation, as well as on techniques for design, test and verification that enable the establishment of local behaviours, such as interactions and tasks, in anticipation of a desired high-level goal.

ESOA'03 workshop was held at the Second Autonomous Agents & Multi-Agent Systems (AAMAS 2003). The post-proceedings including revised versions of the presented papers and additional invited papers are being formally published by Springer-Verlag and they will appear in early 2004.

Topics of Interest
The main topics of interest include but are not restricted to:

- Interaction Mechanisms
Bio-inspired interaction mechanisms
Tag-based interaction mechanisms
Specification-based interaction mechanisms
Trust-based interaction mechanisms

- Middleware technology
Bio-inspired coordination techniques
Self-organisation and indirect coordination
Run-time environment for self-organising agents

- Methodologies, Models, Tools
Methodologies for engineering self-organisation
Analytic models of emergent behaviour
Self-organising theories, methods and algorithms applied to multi-agent systems
Design of agent-based self-organising systems
Design of high-level goals and local interactions
Bio-inspired process algebra and formal specifications
Formal approaches to handling local/global agent behaviour
Models, methods and tools for achieving global coherent behaviour
Self-organising software characterisation frameworks
Self-organising assessment metrics
Approaches to handling robustness, scalability, efficiency and maintenance in self-organising applications
Performance engineering of emergent behaviour in multi agent-based systems
Game theoretic approaches to emergence in multi-agent systems
Cellular automata approaches to emergence in multi-agent systems
Self-organisation to support multi-agent scalability
Challenges in engineering self-organising applications
Self-organising software architectures

- Applications
Self-organising multi-agent based applications
Industrial self-organising applications
Self-organisation in manufacturing systems and supply chain management
Self-organisation in business
Emergent information systems
Self-organising computer networks
Self-organisation in P2P and Grid systems

Submissions
The submission should not exceed 15 pages in the Springer-Verlag LNCS style ( http://www.springer.de/comp/lncs/authors.html ). It must be sent either in PostScript or PDF format to esoa04 at cui.unige.ch. For those submitting papers in PDF format, please ensure that papers do not contain embedded fonts (such as embedded Asian fonts). These papers can be difficult to read and view with many PDF viewers not equipped to display certain character sets.
Publication
All accepted papers will be available on the day of the workshop in a set of working notes. Accepted papers will also made available in electronic format before the day of the workshop. We subsequently aim to publish a common volume collecting the best papers from this workshop and additional contributions, which will be published by Springer Verlag.

Important Dates
Paper submission deadline:

April 1, 2004

Notification of acceptance:

May 1, 2004

Camera ready due:

May 24, 2004

Workshop:

July 20, 2004

Organising Committee
Sven Brueckner, Altarum Institute, USA
http://www.erim.org/~sbrueckner/
Sven.Brueckner at altarum.org

Giovanna Di Marzo Serugendo, University of Geneva, Switzerland
http://cui.unige.ch/~dimarzo
Giovanna.Dimarzo at cui.unige.ch

Anthony Karageorgos, University of Thessaly, Greece
karageorgos at computer.org

Radhika Nagpal, Harvard University, USA
http://www.eecs.harvard.edu/~rad/
rad at eecs.harvard.edu

Steering Committee
H. Van Dyke Parunak, Altarum Institute, USA
Omer F. Rana, University of Cardiff, UK
Franco Zambonelli, University of Modena e Reggio Emilia, Italy

Program Committee
Marco Dorigo, IRIDIA, Université Libre de Bruxelles, Belgium
Noria Foukia, University of Geneva, Switzerland
Nigel Gilbert, University of Surrey, UK
Maria Gini, University of Minnesota, USA
David Hales, University of Bologna, Italy
Salima Hassas, University of Lyon, France
Manfred Hauswirth, Swiss Federal Institute of Technology, Switzerland
Margaret Jefferies, The University of Waikato, New Zealand
Manolis Koubarakis, Technical University of Crete, Greece
Mark Klein, MIT Sloan School of Management, USA
Ghita Kouadri Mostefaoui, University of Fribourg, Switzerland
Soraya Kouadri Mostefaoui, University of Fribourg, Switzerland
Marco Mamei, University of Modena and Reggio Emilia, Italy
Paul Marrow, BT, UK
Philippe Massonet, CETIC, Belgium
Jean-Pierre Mueller, CIRAD, France

N. C. Narendra, Hewlett-Packard, India
Andrea Omicini, University of Bologna, Italy
Daniel Polani, University of Hertfordshire, UK
Martin Purvis, University of Otago, New Zealand
Vitorino Ramos, Istituto Superior Técnico, Lisbon, Portugal
Simon Thompson, BT Exact Technologies, UK
Mihaela Ulieru, University of Calgary, Canada
Paul Valckenaers, Katholieke Universiteit Leuven, Belgium
Chris Van Aart, University of Amsterdam, Netherlands
Tom Wagner, Honeywell Laboratories, USA