First-order learning systems (e.g., FOIL, FOCL, FORTE) generally rely on hill-climbing heuristics in order to avoid the combinatorial explosion inherent in learning first-order concepts. However, hill-climbing leaves these systems vulnerable to local maxima and local plateaus. We present a method, called relational pathfinding, which has proven highly effective in escaping local maxima and crossing local plateaus. We present our algorithm and provide learning results in two domains: family relationships and qualitative model building.

PDF, PS

In Proceedings of the Tenth National Conference on Artificial Intelligence (AAAI-92), pp. 50-55, San Jose, CA, July 1992.

@InProceedings{richards:aaai92, title={Learning Relations by Pathfinding}, author={Bradley L. Richards and Raymond J. Mooney}, booktitle={Proceedings of the Tenth National Conference on Artificial Intelligence (AAAI-92)}, month={July}, address={San Jose, CA}, key={FORTE}, pages={50-55}, url="http://www.cs.utexas.edu/users/ai-lab?richards:aaai92", year={1992} }