R. Maclin & J. Shavlik (1991).
Refining Domain Theories Expressed as Finite-State Automata. Proceedings of the Eighth International Machine Learning Workshop, pp. 524-528, Evanston, IL.

This publication is available in PDF and available in postscript.

Abstract:

The KBANN system uses neural networks to refine domain theories. Currently, domain knowledge in KBANN is expressed as non-recursive, propositional rules. We extend KBANN to domain theories expressed as finite-state automata. We apply finite-state KBANN to the task of predicting how proteins fold, producing a small but statistically significant gain in accuracy over both a standard neural network approach and a non-learning algorithm from the biological literature. Our method shows promise at solving this and other real-world problems that can be described in terms of state-dependent decisions.

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