Generalizing on Multiple Grounds: Performance Learning in Model-Based Troubleshooting

摘要

Model-based reasoning about physical systems has several well-known advantages over heuristic expert systems. These include correctness of conclusions, explanations of conclusions, ease of modifiability and ease of transfer of expertise to new physical systems. On the other hand, reasoning from a model can be slow. This thesis explores ways to augment a model-based diagnostic program with a learning component, so that it speeds up as it solves problems. Several learning components are proposed, each exploiting a different kind of similarity between diagnostic examples. Through analysis and experiments, we explore the effect each learning component has on the performance of a model-based diagnostic program. We also analyze more abstractly the performance effects of Explanation-Based Generalization, a technology that is used in several of the proposed learning components. Keywords: Learning; Explanation based learning; Learning machines; Troubleshooting. (SDW)