Network Structuring and Training Using Rule-based Knowledge
Volker Tresp, Jürgen Hollatz, Subutai Ahmad · 1992
We demonstrate in this paper how certain forms of rule-based knowledge can be used to prestructure a neural network of normalized basis functions and give a probabilistic interpretation of the network architecture. We describe several ways to assure that rule-based knowledge is preserved during training and present a method for complexity reduction that tries to minimize the number of rules and the number of conjuncts. After training, the refined rules are extracted and analyzed. Mail address: Siemens AG, Central Research, Otto-Hahn-Ring 6, 8000 Munchen 83. 1 INTRODUCTION Training a network to model a high dimensional input/output mapping with only a small amount of training data is only possible if the underlying map is of low complexity: a more complex mapping requires a more complex network which results in high parameter variances and, as a consequence, in a high prediction error. This predicament can be solved if we manage to incorporate prior knowledge to bias the network as...