Basic Framework

Abstract

The main purpose of this chapter is to identify the conceptual entities needed to specify a discrete time neural network and to put them together in a formal mathematical structure, i.e. to construct the Ontology for neural networks. This ontology study allows us to achieve the following goals:

• to built a general conceptual structure for neural networks that can cover most of the current models.

• to supply a basic framework for designing, analyzing and implementing neural network architectures.

• to propose a method to calculate the minimum delay required for a neural net to propagate the effects of input signals to the output port.

• to formalize the multi-level adaptation process for neural network systems. In particular, I defined the concept of “Structure Level Neural Networks”, which are networks capable of changing their structures autonomously.