Learning in mathematically-based domains: Understanding and generalizing obstacle cancellations

摘要

Mathematical reasoning provides the basis for problem solving and learning in many complex domains. This paper presents an approach for applying explanation-based learning in mathematically-based domains and describes an implemented learning system based on this approach. In explanation-based learning, a specific problem's solution is generalized into a form that can be later used to solve conceptually similar problems. The manner in which variables are canceled in specific problem solutions guides the presented system's mathematical reasoning process. Analyzing the cancellation of obstacles—variables that preclude the direct evaluation of the problem's unknown—leads to the generalization of the specific solution. Two important general issues in explanation-based learning are also addressed. Namely, generalizing the number of entities in a situation and acquiring efficiently-applicable concepts.