Abstract
We propose a framework that adds learning for improving plan selection in the popular BDI agent programming paradigm. In contrast with previous proposals, the approach given here is able to scale up well with the complexity of the agent’s plan library. Technically, we develop a novel con?dence measure which allows the agent to adjust its reliance on the learning dynamically, facilitating in principle in?nitely many (re)learning phases. We demonstrate the bene?ts of the approach in an example controller for energy management.