Conference on Artificial Intelligence ( IJCAI - 97 ) , pp . 1227 - 1232 , Nagoya , Japan , August 1997 Learning to Improve both E ciency and Quality of Planning

Most research in learning for planning has concentrated on e ciency gains. Another important goal is improving the quality of nal plans. Learning to improve plan quality has been examined by a few researchers, however, little research has been done learning to improve both e ciency and quality. This paper explores this problem by using the Scope learning system to acquire control knowledge that improves on both of these metrics. Since Scope uses a very exible training approach, we can easily focus its learning algorithm to prefer search paths that are better for particular evaluation metrics. Experimental results show that Scope can signi cantly improve both the quality of nal plans and overall planning e ciency.