TITLE:   Learning in asymptotically behaving neural networks
	 
AUTHORS: Dimopoulos, N.J., D. Radvan and W.A. Keddy
	 
IN:      Proceedings of the 1990 International Joint Conference 
	 on Neural Networks, San Diego CA,, 
	 vol. III, pp. 233-238, June 1990. 
	 

ABSTRACT 
The asymptotic behavior of Neural Networks modeled as a set
of nonlinear differential equations of the form
T X + X = W*f(X) + b where X is the neural membrane potential
vector, W is the network connectivity matrix, and f(X) is the 
nonlinearity (an essentially sigmoid function), has been studied
in [4]. 
This behavior depends solely on the topology the network 
as expressed by the connectivity matrix W. 
In this work, we present some results of Hebbian Learning by 
neural networks exhibiting asymptotic behavior as stipulated 
by their connectivity matrices W.
We also present the simulator that has been developed specifically 
for this type of neural networks as well as typical examples.