Requirements on the Use of Goal-Directed Imitation for Self-Adaptation

Imitation is a powerful mechanism in human society, and has proven its usefulness in machine learning. This paper describes the notion of goal-directed imitation and its possible role in realizing self-adaptive systems. The conditions for its application are analyzed. We illustrate the working of imitation in self-adaptation by means of a small example