We can use the function NEWRB to quickly create a radial
 basis network which will approximate the function defined
 by P and T.  The function NEWRB adds neurons to the hidden 
 layer of a radial basis network until it meets the
 specified mean squared error goal.  The spread determines
 the smoothness of the function approximation - a larger
 spread creates a smoother approximation.

 >> eg = 0.02; % sum-squared error goal
 >> sc =1;     % spread constant
 >> net=newrb(P,T,eg,sc);