Multi-Agent Systems

Multi-Agent Systems (MAS) focuses on systems in which intelligent agents interact wich each other. The agents are considered to be autonomous entities that interact with each other in a cooperative or selfisch way. Multi-Agent Systems manifest self-organization and complex behaviors even if the individual strategies of all agents are simple. This leads to a distributed problem solving network where each agent is a semi-autonomous processing node. The results of such a coupled networks of cooperative problem-solver entities are beyond the individual capabilities or knowledge of each entity.

This was showed during the mid-1970s, when researchers form Distributed Artificial Intelligence (DAI) began to formulate some of the basic theories, architectures, and experiments that showed how interaction and division of labor could be effectiveliy applied to problem solving. The experiments revealed intelligent behaviour as an effect that emerges form the interaction of entites with simpler behaviors.

The MAS resarch draw on ideas form many disciplines outside of artificial intelligence, including economics, biology, socilogy, management science, complex systems and philosophy. So the software of a Multi-Agent system operates in environments that are distributed and open, as modern IT infrastructure in todays companies.