RESEARCH ON ELECTRONIC INSTITUTIONS

Human organisations define the roles and responsibilities for organisational participants, who are expected to bring those into action depending on the task and environmental demands. Early work in DAI identified the advantages of organisational structuring as one of the main issues in order to cope with the complexity of designing DAI systems.

However, DAI research, and MAS research in particular, have traditionally kept an individualistic character, and have focused on the principled construction of individual agents following an agent-centered view.

Nonetheless, recently there is a growing interest in incorporating organisational concepts into multi-agent systems, with the purpose of considering organisation-centered designs of multi-agent systems. However, these organisational approaches do not conveniently handle the issues inherent to open multi-agent systems, namely heterogeneity of agents; trust and accountability; exception handling (detection, prevention and recovery from failures that may jeopardise the global operation of the system); and societal change (capability of accommodating structural changes).

Human societies have successfully coped with similar issues, by creating institutions that set and enforce laws, monitor and respond to emergencies, prevent and recover from disasters, etc. In this line of research at the IIIA we propose a similar approach for advocate for the adoption of a mimetic strategy for the realisation of open multi-agent systems founded on the deployment of electronic institutions.

Since we regard electronic institutions as highly complex and critical, we take the view that their design and development must be guided by a principled methodology. In this sense, we advocate for introducing formal methods that underpin the specification, analysis and development of electronic institutions. Thus, in several papers we advocate for a specification formalism for electronic institutions.

The core notions of an electronic institution include:

Agents and Roles. Agents are the players in an electronic institution, interacting by the exchange of illocutions, whereas roles are defined as standardised patterns of behaviour. The identification and regulation of roles is considered as part of the formalisation process of any organisation. Any agent within an electronic institution is required to adopt some role(s). As dialogic actions are associated to roles, an agent adopting a given role is allowed to perform the actions associated to that role. A major advantage of using roles is that they can be updated without having to update the actions for every agent on an individual basis.

Dialogic framework. The context or framework of interaction amongst agents of an institution, such as the objects of the world and the language employed for communicating, are fixed. In a dialogic institution, agents interact through illocutions. Institutions establish the acceptable illocutions by defining the ontology (vocabulary) ---the common language to represent the "world"--- and the common language for communication and knowledge representation. All of these contextual features are bundled together in what we call dialogic framework. By sharing a dialogic framework, we enable heterogeneous agents to exchange knowledge with other agents.

Scene. Interactions between agents are articulated through agent group meetings, which we call scenes, with a well-defined communication protocol. We consider the protocol of a scene to be the possible dialogues agents may have.

Performative structure. Scenes can be connected, composing a network of scenes (the so-called performative structure) which captures the existing relationships among scenes. The specification of a peformative structure contains a description of how the different roles can legally move from scene to scene. A performative structure is to contain the multiple, simultaneous ongoing activities, represented by scenes. Agents within a performative structure may participate in different scenes at the same time with different roles.

Normative Rules. Agent actions in the context of an institution may have consequences that either limit or enlarge its subsequent acting possibilities. Such consequences will impose obligations to the agents and affect its possible paths within the performative structure.


Papers

P. Noriega, C. Sierra; Towards Layered Dialogical Agents . Proceedings of the ECAI'96 Workshop Agents Theories, Architectures and Languages. ATAL'96. Budapest, pp. 69-81. (L.N.A.I. Springer, in press) (IIIA-RR-96-19)

http://www.iiia.csic.es/Publications/1996/Buda.ps.Z

E. Plaza, P. Noriega, C. Sierra: The C0nference Center as an agent-Mediated Institution, 1st International Workshop on Agents in Community Ware, at ICMAS98. pp. 73-82.

Juan A. Rodriguez-Aguilar, Francisco J. Martin, Pablo Noriega, Pere Garcia, Carles Sierra: Towards a Test-bed for trading Agents in Electronic Auction Markets, AICOMM, vol 11, pp 5-19.

Carles Sierra, Pablo Noriega, Institucions Electroniques. Primer Congres Catala d'IA, Tarragona, pp. 125-129. (paper90.ps.gz)

E. Plaza, J. L. Arcos, P. Noriega, C. Sierra: Competing Agents in Agent-Mediated Institutions, Personal Technologies, N 2, pp 1-9.

J. A. Rodriguez-Aguilar, F. F. Martin, P. Garcia, P. Noriega, C. Sierra: Towards a formal specification of Complex Social Structures in Multi-agent Systems. LNAI 1624, pp. 284-300.

M. Esteva, J. A. Rodriguez, J. L. Arcos, C. Sierra, P. Garcia: Formalising Agent Mediated Electronic Institutions, CCIA 00pp. 29-38. (paper111.ps.gz)

M. Esteva, J. A. Rodriguez, C. Sierra, P. garcia, J. L. Arcos: On the formal specifications of electronic institutions, LNAI, (paper118.ps.gz)



Introductory papers