Managing Learning Goals in Strategy-Selection Problems

In case-based reasoning systems, several learning techniques may apply to a given situation. In a failuredriven learning environment, the problems of strategy selection are to choose the best set of learning algorithms or strategies that recover from a processing failure and to use the strategies to modify the system’s background knowledge so that the failure will not repeat in similar future situations. A solution to this problem is to treat learning-strategy selection as a planning problem with its own set of goals. Learning goals, as opposed to ordinary goals, specify desired states in the background knowledge of the learner, rather than desired states in the external environment of the planner. But as with traditional goal-based planners, management and pursuit of these learning goals becomes a central issue in learning. Examples are presented from a multistrategy learning system called Meta-AQUA that combines a case-based approach to learning with nonlinear planning in the knowledge space.