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

Abstract

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.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Feb 01, 1989
Accession Number
ADA207960

Entities

People

  • Paul Resnick

Organizations

  • Massachusetts Institute of Technology

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence
  • Computer Programs
  • Computer Science
  • Computers
  • Cost Reductions
  • Electrical Engineering
  • Encapsulation
  • Expert Systems
  • Failure Mode And Effect Analysis
  • Frequency
  • Generators
  • Hypotheses
  • Language
  • Machine Learning
  • Reasoning
  • Troubleshooting

Fields of Study

  • Computer science

Readers

  • Artificial Intelligence
  • Systems Analysis and Design