Multi-agent Cooperation and Stability - Belief-based Stability in Coalition Formation with Uncertainty - page 1

Chapter 1

Multi-agent Cooperation and Stability

It is through cooperation, rather than
conflict, that your greatest successes
will be derived.

Ralph Charell

1.1 Semi-competitive Multi-agent Systems

We begin this topic with a description of some characteristics of multi-
agent systems.
Distributed intelligent agent systems, or simply called multi-agent
systems, have been an area of active research in computer science
during the past few decades. Although the term 'intelligent agent'
itself is not clearly defined, there are some commonly agreed upon
characteristics of intelligent agents, with the most important of which
being autonomy , social ability ,and rationality [1].

1.1.1 Characteristics of Multi-agent Systems

There is no universal agreed upon definition of 'software agent'. How-
ever, it is agreed that, in order for a piece of program to qualify as
an agent, it must have certain properties, and the most important
of which is autonomy . Indeed, all software agents must be highly au-
tonomous , meaning that they can be viewed as independent entities