Towards a Framework for Memory-Based Reasoning

In this paper, we propose a broad framework for Memory-Based Reasoning (MBR). The framework allows us to clarify the technical merits and limitations of several current MBR methods and to design new variants. The proposed computational framework consists of three components: a speciication language to deene an adaptive notion of relevant context for a query; mechanisms for retrieving this context; and local learning procedures that are used to induce the desired action from this context. Based on the framework we derive several analytical and empirical results that shed light on MBR algorithms. We introduce the notion of an MBR transform, and discuss its utility for learning algorithms. We also provide several perspectives on memory-based reasoning from a multidisciplinary point of view.