Title :
Universal approximation using dynamic recurrent neural networks: discrete-time version
Author :
Jin, Liang ; Gupta, Madan M. ; Nikiforuk, Peter N.
Author_Institution :
Coll. of Eng., Saskatchewan Univ., Saskatoon, Sask., Canada
Abstract :
In this paper, the approximation capability of a class of discrete-time dynamic recurrent neural networks (DRNNs) is studied. Analytical results presented show that some of the states of such a DRNN described by a set of difference equations may be used to approximate uniformly a state space trajectory produced by either a discrete-time nonlinear system or a continuous function on a closed discrete-time interval
Keywords :
difference equations; discrete time systems; function approximation; nonlinear systems; recurrent neural nets; difference equations; discrete-time nonlinear system; discrete-time recurrent neural networks; dynamic recurrent neural networks; function approximation; state space trajectory; Difference equations; Feedforward neural networks; Intelligent networks; Intelligent systems; Neural networks; Nonlinear dynamical systems; Nonlinear systems; Recurrent neural networks; State-space methods; Tin;
Conference_Titel :
Neural Networks, 1995. Proceedings., IEEE International Conference on
Conference_Location :
Perth, WA
Print_ISBN :
0-7803-2768-3
DOI :
10.1109/ICNN.1995.488134
