Situated Learning in Autonomous Agents

In Chapter 17 we started the discussion about autonomous agents as part of the situated cognition approach in AI, and we develop this further here focussing on learning. An agent is autonomous when it is able to function on its own in a complex and changing environment. We will call an agent adaptive when, in addition, it is able to improve its behaviour to make it (more) appropriate for its environment and task. Through interactions with the environment, the agent learns to improve its behaviour on the basis of its own appraisal of the perceived external or internal effects of its actions. We will thus focus on learning as a form of adaptation, where learning is an integral part of the agent's behaviour that cannot be disconnected from its performance, as opposed to a separate cognitive function that can be studied and modelled on a stand-alone basis.