Explanation-Based Failure Recovery

Interactions are inherent in design-type problemsolving tasks where only partially compiled operators are available. Failures arising from such interactions can best be recovered by explaining them in the underlying domain models. In this paper we explain how Explanation-Based Learning provides a framework for recovering in this manner. This approach also alleviates some of the problems associated with the least-commitment approach to design-type problem-solving.

Authors

• Ajay Gupta

Keywords

No keywords are indexed for this paper.