Abstract

This paper presents a study of Artificial Neural Nets learning of high level rules. We defined high level rules as symbolic rules that can represent strong relations among their premise components. This kind of rules can model the modulation of some inputs upon the others and context influences. In order to study the practical application of these rules we choose one well known cognitive modelling problem: the Balance Scale Problem. Our simulation results demonstrate the importance of using high-order nets to learn high level rules. 1. INTRODUCTION The development of Artificial Intelligence research (AI) has followed mainly two directions: the use of symbolic AI methods and the use of sub-symbolic connectionist methods (Artificial Neural Networks - ANN). In this framework we have studied different ways to integrate these two approaches through Hybrid NeuroSymbolic Systems. Hybrid neuro-symbolic systems, like KBANN [Towell-91], SYNHESYS [Giacometi-92] or INSS [Osrio-95], gener...

Citations