Mj. Healy et Tp. Caudell, ACQUIRING RULE SETS AS A PRODUCT OF LEARNING IN A LOGICAL NEURAL ARCHITECTURE, IEEE transactions on neural networks, 8(3), 1997, pp. 461-474
Envisioning neural networks as systems that learn rules calls forth th
e verification issues already being studied in knowledge-based systems
engineering, and complicates these with neural-network concepts such
as nonlinear dynamics and distributed memories, We show that the issue
s can be clarified and the learned rules visualized symbolically by fo
rmalizing the semantics of rule-learning in the mathematical language
of two-valued predicate logic, We further show that this can, at least
in some cases, be done with a fairly simple logical model, We illustr
ate this,vith a combination of two example neural-network architecture
s, LAPART, designed to learn rules as logical inferences from binary d
ata patterns, and the stack interval network, which converts real-valu
ed data into binary patterns that preserve the semantics of the orderi
ng of real values, We discuss the significance of the formal model in
facilitating the analysis of the underlying logic of rule-learning and
numerical data representation, We provide examples to illustrate the
formal model, with the combined stack interval/LAPART networks extract
ing rules from numerical data.