What is "THE STANDARD UNIVERSAL APPROXIMATOR NEURAL NETWORK "

The "STANDARD" UNIVERSAL APPROXIMATOR NEURAL NETWORK is a hidden layer regression net with

    1. A SINGLE hidden layer 
    2. A SINGLE bounded nonlinear hidden layer transfer function
    3. A LINEAR output layer transfer function

I'm stating this because it is obvious that some believe that, for a universal approximator:

    1. There has to be more than one hidden layer
 and/or
    2. The output layer transfer funtion has to be nonlinear.

Of course there are additional conditions on function finiteness, etc which I have omitted, but I think I have made my point.

Hope this Helps,

Greg

Answer 1

[Greg Heath]

The question is the answer