Learning in multi-agent systems

Tools

Alonso, Eduardo; d'Inverno, Mark; Kudenko, Daniel; Luck, Michael and Noble, Jason. 2001. Learning in multi-agent systems. The Knowledge Engineering Review, 16(3), pp. 277-284. ISSN 0269-8889 [Article]

Preview

Text (Learning in multi-agent systems)
Learning in multi-agent systems.pdf - Published Version
Download (42kB) | Preview

Abstract or Description

In recent years, multi-agent systems (MASs) have received increasing attention in the artificial
intelligence community. Research in multi-agent systems involves the investigation of autonomous,
rational and flexible behaviour of entities such as software programs or robots, and their interaction and
coordination in such diverse areas as robotics (Kitano et al., 1997), information retrieval and
management (Klusch, 1999), and simulation (Gilbert & Conte, 1995). When designing agent systems,
it is impossible to foresee all the potential situations an agent may encounter and specify an agent
behaviour optimally in advance. Agents therefore have to learn from, and adapt to, their environment,
especially in a multi-agent setting.
In this panel report, we combine several different perspectives, and review some key contributing
influences. The report begins with a discussion of just why learning is considered by many to be a
crucial characteristic of intelligent agent systems. In the following section, the features of different
learning algorithms, and their potential impact on multi-agent systems, are discussed, such as ways of
achieving multi-agent learning, the applicability of off-line learning methods and a discussion of the
pros and cons of reactive, logic-based and social learning methods.