Thank you to all authors who have submitted papers to ESAW'04...
The deadline to submit the final version of the accepted papers (including modifications requested by the reviewers) is 3 October 2004. Send your paper, by e-mail, to Marie-Pierre Gleizes.
Submission format: Contributions should not exceed 12 pages and should be formatted according to the LNCS/LNAI style guide available here.
You can register to the workshop via this page.
We have received 46 papers and we have accepted 25 papers for presentation at the workshop.
Programme
The final programme is downloadable in color and black and white.
Every presentation slot includes 20 minutes for the talk and 5 minutes for questions closely related to the particular presentation (e.g. to clarify understanding). Any ideas for discussion of a wider/broader kind should be taken down and brought up in that session's wrap-up slot.
To enhance the scientific experience of each participant, each author should read a specific paper of his session before the workshop so as to prepare some relevant questions and leads for the session discussion. The aim is not to evaluate the paper or be too critical, but rather to explore more in depth each session's problematic.
| Wednesday Thursday Friday | |||||
| Wednesday 20th October, 2004 | |||||
|---|---|---|---|---|---|
| Time | |||||
| 09:00-09:45 | Local Registration and Coffee Time | ||||
| 09:45-10:00 | Welcome | ||||
| Software Engineering | 10:00-10:25 | A Sample Application of ADELFE Focusing on Analysis and Design G. Picard, D. Capera, M.-P. Gleizes, P. Glize (France) |
 |
 |
|
| 10:25-10:50 | Deployment of Distributed Multi-Agent Systems L. Brauchbach, A. Pokahr, K.-H. Krempels, W. Lamersdorf (Germany) |
|
|
||
| 10:50-11:15 | T-Compound Interaction and Listening Agents
E. Platon, N. Sabouret, S. Honiden (Japan) |
|
|
||
| 11:15-11:40 | SONIA: A Methodology for Natural Agent Development F. Alonso, S. Frutos, L. A. Martínez, C. Montes (Spain) |
|
|
||
| 11:40-12:10 | Discussion | ||||
| 12:10-14:10 | Lunch | ||||
| Negotiation | 14:10-14:35 |
Motivation-Based Selection of Negotiation Partners S. Munroe, M. Luck (United Kingdom) |
|
|
|
| 14:35-15:00 |
Welfare Engineering in Practice: on the Varieties of Multiagent Resource Allocation Problems Y. Chevaleyre, U. Endriss, S. Estivie, N. Maudet (France) |
|
|
||
| 15:00-15:25 |
Managing Conflicts between Individuals and Societies in Multi-Agent Systems R. Fuentes, J. Gómez-Sanz, J. Pavón (Spain) |
|
|
||
| 15:25-15:55 | Discussion | ||||
| 15:55-16:25 | Coffee Break | ||||
| Large Scale | 16:25-16:50 | Semantically Federating Multi-Agent Organizations R. Cenk Erdur, O. Dikenelli, I. Seylan, O. Gurcan (Turkey) |
|
|
|
| 16:50-17:15 |
Using Stand-In Agents in Partially Accessible Multi-Agent Environment M. Rehak, M. Pechoucek, J. Tozicka, D. Sislak (Czech Republic) |
|
|
||
| 17:15-17:30 | Discussion | ||||
| 18:00 | Cocktail | ||||
| |||||
| Thursday 21st October, 2004 | |||||
| Time | |||||
| 08:30-09:30 | Invited speaker: Vincent Chevrier From Self-organized Models to Collective Problem Solving |
||||
| 09:30-09:50 | Coffee Break | ||||
| Roles | 09:50-10:15 |
Achieving Competence by Argumentation on Rules for Roles I. A. Letia, C. Feier, M. Acalovschi (Romania) |
|
|
|
| 10:15-10:40 |
Participation Components for Holding Roles in MAS Protocols C. Sibertin-Blanc, N. Hameurlain (France) |
|
|
||
| 10:40-11:00 | Discussion | ||||
| Organizations | 11:00-11:25 |
Virtual Enterprise Normative Framework within Electronic Institutions H. Lopes Cardoso, E. Oliveira (Portugal) |
|
|
|
| 11:25-11:50 |
Virtual Knowledge Communities for Corporate Knowledge Issues P. Maret, M. Hammond, J. Calmet (France) |
|
|
||
| 11:50-12:00 | Discussion | ||||
| 12:00-14:00 | Lunch | ||||
| Social Aspect | 14:00-14:25 |
Using Social Power to Enable Agents to Reason about Being Part of a Group C. Carabelea, O. Boissier, C. Castelfranchi (France) |
|
|
|
| 14:25-14:50 |
Modelling Flexible Social Commitments and their Enforcement P. Pasquier, R. Flores, B. Chaib-draa (Canada) |
|
|
||
| 14:50-15:15 |
DIAGAL: a Generic ACL for Open Systems P. Pasquier, M. Bergeron, B. Chaib-draa (Canada) |
|
|
||
| 15:15-15:45 | Discussion | ||||
| 15:45-16:15 | Coffee Break | ||||
| Simulation | 16:15-16:40 |
Controlled Experimentation with Agents - Models and Implementations M. Röhl, A. M. Uhrmacher (Germany) |
|
|
|
| 16:40-17:05 |
Techniques for Analysis and Calibration of Multi-Agent Simulations M. Fehler, F. Kluegl, F. Puppe (Germany) |
|
|
||
| 17:05-17:20 | Discussion | ||||
| 19:30 | Social Event (at the Hôtel d'Assezat which hosts the Fondation Bemberg) | ||||
|
| |||||
| Friday 22nd October, 2004 | |||||
| Time | |||||
| 08:30-09:30 | Invited speaker: Pablo Noriega Extending Electronic Institutions - An Explorer's Log |
||||
| 09:30-09:50 | Coffee Break | ||||
| Cooperation | 9:50-10:15 |
Strategies for Distributing Goals in a Team of Cooperative Agents L. Cholvy, C. Garion (France) |
|
|
|
| 10:15-10:40 |
Trust-based Contract Nets R. Falcone, G. Pezzulo, C. Castelfranchi, G. Calvi (Italy) |
|
|
||
| 10:40-11:05 |
Collectively Cognitive Agents in Cooperative Teams J. Brzezinski, P. Dunin-Keplicz, B. Dunin-Keplicz (Poland) |
|
|
||
| 11:05-11:30 |
Cooperative Agent Model Instantiation to Collective Robotics G. Picard |
|
|
||
| 11:30-12:10 | Discussion | ||||
| 12:10-14:10 | Lunch | ||||
| 14:10-14:35 | Stable Multi-Agent Systems A. Bracciali, P. Mancarella, Ki. Stathis, F. Toni (Italy) |
|
|
||
| 14:35-15:00 |
Organizations as Socially Constructed Agents in the Agent Oriented Paradigm G. Boella, L. van der Torre (Italy) |
|
|
||
| 15:00-15:15 | Discussion | ||||
| 15:15-15:45 | Closing of the Workshop | ||||
| 15:45-16:30 | Coffee Time | ||||
|
| |||||
Abstracts
DIAGAL: a Generic ACL for Open Systems
Philippe Pasquier, Mathieu Bergeron, Brahim Chaib-draa (Canada)
Recently, social commitment based approaches have been proposed to solve problems issuing from previous mentalistic based semantics for agent communication languages. In this paper, we present the latest version of our dialogue games based agent communication language (DIAGAL) which allows the agents to manipulate the public layer of social commitments through dialogue. We show that DIAGAL is complete according to the sequential creation, cancellation, update and discharge of social commitments. We also extend and refine notions of success and satisfaction previously associated with speech-acts to this new dialogical setting. Finally, we explain why DIAGAL is a good candidate for open and heterogeneous MAS development.
A Sample Application of ADELFE Focusing on Analysis and Design
Gauthier Picard, Davy Capera, Marie-Pierre Gleizes, Pierre Glize (France)
This paper aims at explaining how to follow an agent-oriented process to develop a multi-agent mechanism design system. ADELFE methodology is devoted to adaptive multi-agent systems in which adaptation is enabled by cooperative self-organization. Two main works are emphasized. First, the analysis leads to the agent identification by studying the interactions both between the system and its environment and within the system itself. Second, the different modules of the agents and their cooperative attitude are modeled during the design phase. Such an approach is promising, but raises some difficulties considering the notion of cooperation, which is discussed before concluding.
Strategies for Distributing Goals in a Team of Cooperative Agents
Laurence Cholvy, Christophe Garion (France)
This paper addresses the problem of distributing goals to individual agents inside a team of cooperative agents. It shows that several parameters determine the goals of particular agents. The first parameter is the set of goals allocated to the team; the second parameter is the description of the real actual world; the third parameter is the description of the agents' ability and commitments. The last parameter is the strategy the team agrees on: for each precise goal, the team may define several strategies which are orders between agents representing, for instance, their relative competence or their relative cost. This paper also shows how to combine strategies. The method used here assumes an order of priority between strategies.
Collectively Cognitive Agents in Cooperative Teams
Jacek Brzezinski, Piotr Dunin-Keplicz, Barbara Dunin-Keplicz (Poland)
This research continues a line of recent investigation resulting already in Dunin-Keplicz and Verbrugge theory of collective motivational attitudes as well as a formal theory of teamwork. In this paper we aim to describe our work over a theory of collective commitments in cooperative teams basing on a software test-bed for conducting trust-based agent experiments. First, short introductions to the theories of collective commitments and trust are given. Next, the most important properties of the system are presented together with a scenario of interplay. Finally several tests are described that compare different versions of a commitment applied in various situations.
Participation Components for Holding Roles in MAS Protocols
Christophe Sibertin-Blanc, Nabil Hameurlain (France)
An autonomous agent in a MAS involves in a protocol - more exactly in a conversation following the rules of a protocol - in order to reach objectives, some ones shared with all participants, some others specific and private. We assume each conversation is monitored by a middleware component - the conversation's moderator - that guarantees that the shared objectives will be reached. This paper addresses the means needed by an agent to be able to reach its own objectives in the course of a conversation. The first step is to identify these objectives and thus to distinguish the strategic and tactic levels in agents' behaviours. The strategic level must be handled by the agent itself; the required capabilities are abstract and relevant for larges categories of similar protocols. Once a strategy is set for a conversation, its application at the tactic level can be delegated to a middleware component, called a participation, that intervenes in the conversation on the behalf of the agent. This component is specific to the role held by the agent and it is tailored to make the best use of the subtleties of the protocol’s rule. This approach brings many engineering benefits.
Achieving Competence by Argumentation on Rules for Roles
Ioan Alfred Letia, Cristina Feier, Monica Acalovschi (Romania)
We consider the deep venous thrombosis (DVT) as a case study for the specification and implementation of a multi-agent system. The DVT is an application with low clinical accuracy, needing objective tests, some of them satisfactorily accurate in experienced hands and others more definite but invasive. Whether one or more decision makers are involved in this activity is a matter of context, but the main events are decided by a process that has in itself some forms of argumentation. Our approach is an argumentative multi-agent system specified by rules capturing various roles in the diagnosis activity. Although the DVT scenario is a real one, more aspects of helath care than the ones presented in this paper can conveniently accomodated in this framework by extending the set of roles and refining the set of rules.
The Enforcement of Flexible Social Commitments
Philippe Pasquier, Roberto Flores, Brahim Chaib-draa (Canada)
For over a decade, agent research has shown that social commitments support the definition of open multiagent systems by capturing the responsibilities that agents contract toward one another through their communications. These systems, however, rely on the assumption that agents respect the social commitments they adopt. To overcome this limitation, in this paper we investigate the role of sanctions as elements whose enforcement fosters agents compliance with adopted commitments. In particular, we present a model of social commitments to which sanctions are attached, and where the enforcement of sanctions act as a social control mechanism for the satisfaction of commitments.
Stable Multi-agent Systems
Andrea Bracciali, Paolo Mancarella, Kostas Stathis, Francesca Toni (Italy)
We present an abstract declarative semantics for multi-agent systems based on the idea of {\em stable set}, and argue that it can be suitably employed to describe, and to some extent verify, the dynamics of complex systems of autonomous and heterogeneous interacting agents. We view agents as black-boxes, whose semantics is abstractly understood as an input-output transformation from the agents' observations about their environment, to the actions they perform. Stable sets (of observations and actions) characterise multi-agent systems able to reach an equilibrium point. Our semantics via stable sets takes into account the possibility that agents may fail. We illustrate how stability can characterise multi-agent systems by means of examples. We also draw considerations about how stable sets can be effectively approximated.
SONIA: A Methodology for Natural Agent Development
Fernando Alonso, Sonia Frutos, Loïc A. Martínez, César Montes (Spain)
Agent-Oriented Software Engineering has emerged as a powerful engineering discipline that can deal with the complexity of today's software systems (primarily in distributed and open environments) better than other more traditional approaches. However, AOSE does not provide a software development process that naturally leads, if the problem so requires, to an agent architecture or, where appropriate, an agent society. In current agent development methodologies, either the development processes target an agent organization, which is not necessarily always the best structure, as of the requirements definition stage, or the identification and design of agents are complex, and designer experience plays an essential role in their definition. In this paper, we present the SONIA methodology (Set of mOdels for a Natural Identification of Agents) in an attempt to solve this problem. Based on a generic problem-independent analysis and a bottom-up development process, SONIA naturally outputs an agent-based system and, where appropriate, an agent society design.
Deployment of Distributed Multi-Agent Systems
Lars Braubach, Alexander Pokahr, Karl-Heinz Krempels, Winfried Lamersdorf (Germany)
The agent metaphor has shown its usefulness for modelling as well as implementing complex and dynamic applications. Although a number of agent applications has been successfully realised and used, it must be stated that the distribution of commercial off-the-shelf applications is very scarce. For this discontenting situation, at least two reasons can be identified. On the one hand, the development of agent based applications is difficult suffering from insufficient standards and tools and on the other hand deployment issues are little researched and supported. In this paper, several deployment related topics are discussed and a vision for the deployment of distributed multi-agent systems is conceived. From the vision, requirements for launching and configuring agent applications are derived. According to these requirements, a platform independent reference model of the proposed deployment infrastructure is presented. The reference model provides the basis for the development of our ASCML (Agent Society Configuration Manager and Launcher) tool, which is currently implemented for the JADE and Jadex multi-agent platforms.
Controlled Experimentation with Agents - Models and Implementations
Mathias Röhl, Adelinde M. Uhrmacher (Germany)
The deployment of multi-agent systems demands for justified confidence into their behaviour, both with respect to correct results of computations and with respect to timeliness thereof. Depending on the stage of the development process different mechanisms and abstractions are needed to facilitate the rigorous evaluation of interacting agents. We propose a modelling and simulation framework for supporting the development process of multi-agent systems; from specifications to implementations. Therefore, experimental set-ups are needed that allow an incremental refinement of agents while providing rigorous observation facilities. The benefit of using modelling and simulation for evaluating cooperative agents is illustrated using a simple example based on the Contract Net Protocol.
Managing Conflicts between Individuals and Societies in Multi-Agent Systems
Rubén Fuentes, Jorge Gómez-Sanz, Juan Pavón (Spain)
The development of multi-agent systems (MAS) implies considering both the social and individual levels of these systems. However, the elements in these levels are not necessarily consistent. Conflicts can arise between the goals of the community and those of individual agents. These contradictions are potentially very complex, given the inherent intentional nature of agents and the interactive features of their societies. Developers can face these problems about contradictions with traditional software engineering verification techniques. Nevertheless, these techniques always depend on the understanding of developers about what properties to check. Abstractions in the agent paradigm offer a new possibility for verification: the use of expert knowledge from social sciences to detect and solve this kind of problems. Social sciences and MAS research share a similar view about their objects of study: societies of intentional actors. This allow the adaptation of the expertise of social sciences to the study of MAS, providing a new source of knowledge for the verification of MAS focused in their social and intentional features. Based in our previous research, we have developed a method to solve motivational contradictions with one of these social theories, the Activity Theory. This paper gives an overview of the verification method with this theory and applies it to a case study about libraries in Internet.
Techniques for Analysis and Calibration of Multi-Agent Simulations
Manuel Fehler, Franziska Kluegl, Frank Puppe (Germany)
In this paper we present analysis and calibration techniques that exploit knowledge about a multi agent society in order to calibrate the system parameters of a corresponding society simulation model. The techniques address typical problems of multi agent simulation calibration like the vast amount of parameters that need to be calibrated, the complex parameter dependencies due to interactions between the simulated agents and the generally enormous computational cost of running a multi agent simulation in order to overcome the problems of standard society calibration techniques, which simply apply search heuristics to the parameter search space.
Using Social Power to Enable Agents to Reason about Being Part of a Group
Cosmin Carabelea, Olivier Boissier, Cristiano Castelfranchi (France)
One of the main challenges in multi-agent systems is the coordination of autonomous agents. In order to achieve this coordination, the agents are considered to be part of what we call a group (e.g., organization, institution, team, normative society, etc.). Our goal is to enable an agent to reason about the implications of being part of a group: what does it gain or lose, what are the constraints imposed on its behaviour. The theory of social power has been proposed as a paradigm to describe the agent's behaviour. In this paper we use this theory, we formalize it and we extend it to include group-related aspects. We then show how, using this theory, an agent is able to reason about the constraints imposed on its behaviour by the group, for example to decide whether it should enter or not a group.
Cooperative Agent Model Instantiation to Collective Robotics
Gauthier Picard (France)
The general aim of our work is to provide tools, methods and models to adaptive multi-agent systems designers. These systems consist in several interacting agents and have to optimize problem solving in a dynamic environment. In this context, ADELFE method, which is based on a self-organizing adaptive multi-agent system model, was developed. Cooperation is used as a local criterion to self-organize the collective in order to reach functional adequacy with the environment. One key stage during the design process is to instantiate a cooperative agent model that is an extension to classical reactive models in which cooperation subsumes any other nominal behavior. A sample implementation of the agent model in the collective robotics domain - resource transportation - will illustrate a discussion on the model.
Virtual Enterprise Normative Framework within Electronic Institutions
Henrique Lopes Cardoso, Eugénio Oliveira (Portugal)
Virtual Enterprises are a major trend within the B2B scenario. Technological support towards enabling this cooperation model includes the multi-agent systems paradigm. In this paper we identify requirements of Virtual Enterprise contracts, developing a normative framework for contract validation and enforcement. Furthermore, we enclose this conception within the structure of an Electronic Institution, which governs and supports the interaction of agents in business scenarios, providing specific services such as brokering, reputation, negotiation mediation, and contract related services. We focus on electronic contracting as a means of establishing cooperation agreements, and we describe the institution’s role on the e-contracting life-cycle.
Trust-based Contract Nets
Rino Falcone, Giovanni Pezzulo, Cristiano Castelfranchi, Gianguglielmo Calvi (Italy)
In this paper we use a contract net protocol in order to compare various delegation strategies. We have implemented some different agent, having a set of tasks to delegate (or to perform by themselves); the tasks are performed by the agents in a dynamic environment, that can help or worse their activity. The agent rely upon different strategies in order to choose whom to delegate. We implemented three classes of trustiers: a random trustier (who randomly chooses the trustee whom delegate the task to; this is used as baseline); a statistical trustier (who builds the trustworthiness of other agents only on the basis of their previous performances, without considering specific features of these agents and without considering the environment in which they performed); a cognitive trustier (who builds a sophisticated and cognitively motivated trust model of the trustee, taking into account its specific features, its ability and motivational disposition, and the impact of the environment on its performance). Our experiments show the advantage of using cognitive representations: the cognitive trustier always wins.
Semantically Federating Multi-Agent Organizations
Riza Cenk Erdur, Oguz Dikenelli, Inanc Seylan, Onder Gurcan (Turkey)
Based on the progress on multi-agent technologies, we can expect the existence of a large number of multi-agent organizations operating in specific domains. Hence, the semantic interoperability of these multi-agent organizations will gain more importance. In this paper, first a conceptual architecture for the semantic interoperability of multi-agent organizations is introduced. Then, ontological infrastructure of this conceptual architecture is discussed. Finally, implementation details are given focusing on the collaboration of the architectural elements and semantic matching process.
Motivation-Based Selection of Negotiation Partners
Steve Munroe, Michael Luck (United Kingdom)
As agent systems become more complex, there is a growing need to equip agents with mechanisms that enable greater degrees of autonomy in order that such complexity can be managed. This is especially the case when it comes to agent-to-agent interaction which, in systems of selfish agents, often follows the format of negotiation. One particular problem which has hitherto received little attention is the problem of identifying appropriate negotiation partners. The problem is particularly difficult in dynamic systems where the rewards gained from having a goal satisfied, and the resources used to satisfy goals, change over time. Such dynamics demands higher degrees of autonomy from agents so that such factors can be handled at run-time and without the aid of human controllers. To that end, this paper draws inspiration from biological organisms and theories of motivation, and describes a motivation-based mechanism to evaluate and select between negotiation candidates. Acceptable candidates are identified using motivation-based thresholds on objective scoring measures, then, the importance of negotiation issues is considered, determined by motivations, and expected performance measures are evaluated accordingly. The mechanism is described and empirically evaluated.
Welfare Engineering in Practice: on the Varieties of Multiagent Resource Allocation Problems
Yann Chevaleyre, Ulle Endriss, Sylvia Estivie, Nicolas Maudet (France)
Many multiagent applications can be abstractely regarded as distributed resource allocation problems. By doing so, we can envisage to borrow results known to be true for any instances of this problem, in order to welfare engineer a society opf agents. How precisely such a mapping can be achieved, however, has not been studied so far. In this paper, we exhibit a tentative list of criteria that will help the designer in this task, and illustrate on several examples how this can be done in practice.
Organizations as Socially Constructed Agents in the Agent Oriented Paradigm
Guido Boella, Leendert van der Torre (Italy)
In this paper we propose a new role for the agent metaphor in the definition of the organizational structure of multiagent systems. The agent metaphor is extended to consider as agents also social entities like organizations, groups and normative systems, so that mental attitudes can be attributed to them - beliefs, desires and goals - and also an autonomous and proactive behavior. We show how the metaphor can be applied also to structure organizations in functional areas and roles, which are described as agents too. Thus, the agent metaphor can play a role similar to the object oriented metaphor which allows structuring objects in component objects. Finally, we discuss how the agent metaphor addresses the problems of control and communication in such structured organizations.
T-shaped Interaction Compounds and Listening Agents
Eric Platon, Nicolas Sabouret, Shinichi Honiden (Japan)
Listening agents represent a natural and common situation of persons influenced by the discussion they are listening to. Side-effects of listening without being directly involved in the discussion are various and may be relevant in multi-agent systems. We present in this paper the T-model that synthesises such situations and may serve as a building block for software engineers to design advanced agent societies. Foundations of a formal model are detailed and the recurrence of the T-shape interaction is highlighted in well-known past projects.
Large-Scale Distributed Directory Structures: A Challenge Problem for Social Multi-Agent Systems?
David Cabanillas, Steven Willmott (Spain)
Directory services are an integral part of today’s computer networking technology – underpinning Internet naming systems and are increasingly being extended to hold rich structured data for automated service discovery in new generations of Semantic Web and Web Services systems. Directory services also represent some of the largest electronic structured / semi-structured artificial social systems to date. In this paper we argue that as directory services increasingly become part of our computing infrastructure grow in sophistication techniques from multi-agent systems organization theory, social simulation and related fields will become increasingly important in understanding the resulting systems. Further we suggest that the community might discuss drawing up commonly generic models of directory structures which could be used as challenge problems to focus the application of technology. The paper includes a simple draft model and discussion relevant fields as well as how commonly agreed generic models could be drawn up.
Agent Societies for Corporate Knowledge Issues
Pierre Maret, Mark Hammond, Jacques Calmet (France)
Corporate knowledge consists both of information that is available throughout a company and of information technology frameworks and paradigms. Considering an enterprise as a distributed computational paradigm, multi-agent systems can be proposed to address knowledge management issues within a company. We consider in this paper a new approach for corporate knowledge based on the agent oriented abstraction paradigm. This paradigm provides a high level of abstraction. We investigate here the concept of virtual knowledge communities, which is a convenient concept for addressing dynamical distributed knowledge management. It allows improved simulation and support for knowledge management processes, and therefore to innovate with new methods in this field. Our approach is well-suited for instance to filter the amount of knowledge that is transmitted throughout a company.
Using Stand-In Agents in Partially Accessible Multi-Agent Environment
Martin Rehak, Michal Pechoucek, Jan Tozicka, David Sislak (Czech Republic)
This contribution defines a metrics and proposes a solution for the problem of agents inaccessibility in multi-agent systems. We define the stand-in pattern for knowledge maintenance in distributed agent systems with inaccessibility. Our implementation has been designed and tested in the A-globe agent platform. We also present a set of measurements quantifying agents’ inaccessibility in our domain and comparing the usefulness of different inaccessibility solution methods as a function of overall accessibility.
