Norms and Agents

Trackchairs

Abstract

Multi-Agent models and technologies contribute to the development of communication and cooperation infrastructure for decentralized and open applications (e.g. e-commerce, interactive games, social and collective robotics, ambient intelligence...) They also contribute to the simulation of collective phenomena in order to explain and understand their complex behaviours.

Although autonomy is a central property of this paradigm, control and regulation are important topics. Autonomous agents, i.e.digital entities taking part in the achievement of complex tasks in interaction with other agents or humans, must be controlled and regulated in order to exhibit coherent behaviours at the global system level. Research in the domain have thus proposed different models aiming at this regulation: organizations, policies, norms, institutions... The current evolution of the applications stresses the requirements of such regulation mechanisms while preserving autonomy, openness and decentralization.

We aim at reviewing the state of the art of the research in the multi-agent domain, and more generally in Artificial Intelligence, as far as the following topics are concerned: