G. Towell & J. Shavlik (1994).
Knowledge-Based Artificial Neural Networks. Artificial Intelligence, 70, pp. 119-165.

This publication is available in PDF and available in postscript.

Abstract:

Explanation-based and empirical learning are two largely complementary methods of machine learning. These approaches both have serious problems which preclude their being a general-purpose learning method. However, a ``hybrid'' learning method that combines explanation-based with empirical learning may be able to use the strengths of one learning method to address the weaknesses of the other, thereby resulting in a system superior to either approach in isolation. KBANN (Knowledge-Based Artificial Neural Networks) is a hybrid learning system built on top of connectionist learning techniques. It maps problem-specific ``domain theories'', represented in propositional logic, into neural networks and then refines this reformulated knowledge using backpropagation. KBANN is evaluated by extensive empirical tests in the domain of molecular biology. Among other results, these tests show that the networks created by KBANN are superior, in terms of their ability to correctly classify unseen examples, to a wide variety of learning systems, as well as techniques proposed by biologists.

Return to the publications of the Univ. of Wisconsin Machine Learning Research Group.