Dynamic Neural Networlks: An Overview

Over the last decade several advance< havc been made in the pddigm of artificial neurd networks with specific emphasis on architectures ad learning algorithms. However, most of the work is fixu.stxi on static ( f d o n m d ) neural networks. These neural networks respond instantaneously to the inputs, for they do not posses any time delay units. The use of time delays in neural networks is neurobiologically motivated, since it is well k m w n that signal delays arc omnipresent in the brain aal play an important role in neurobiological information processing. This conccpl has led to the development of dynamic neural networks. It is envisaged that dynamic neural networks, in addition to better represcntation of biological neural systems, offer better L omputationdl capabilities compared to their static counterparts. The objective of this paptx is to give an overview of dynamic neural structures.