Proceedings of the Third International Workshop on Multi-Strategy Learning (MSL-96), pp. 271-279, Harpers Ferry, WV, May 1996.
Paper ID: 60
Category: Inductive Logic Programming, Learning for Planning and Problem Solving, Explanation-Based Learning
Most approaches to learning control information in planning systems use explanation-based learning to generate control rules. Unfortunately, EBL alone often produces overly complex rules that actually decrease planning efficiency. This paper presents a novel learning approach for control knowledge acquisition that integrates explanation-based learning with techniques from inductive logic programming. EBL is used to constrain an inductive search for selection heuristics that help a planner choose between competing plan refinements. SCOPE is one of the few systems to address learning control information in the newer partial-order planners. Specifically, SCOPE learns domain-specific control rules for a version of the UCPOP planning algorithm. The resulting system is shown to produce significant speedup in two different planning domains.
