Rule-based Analysis of Behaviour Learned by Evolutionary and Reinforcement Algorithms

Abstract

We study behavioural patterns learned by a robotic agent by means of two different control and adaptive approaches - a radial basis function neural network trained by evolutionary algorithm, and a traditional reinforcement Q-learning algorithm. In both cases, a set of rules controlling the agent is derived from the learned controllers, and these sets are compared. It is shown that both procedures lead to reasonable and compact, albeit rather different, rule sets.

Categories and tags

  • Konferenční příspěvek (mezinárodní konf.)