For an ART2 network the weights of the top-down-links (F F links) are set to 0.0 according to the theory ([CG87b]).

The choice of the initial bottom-up-weights is determined as follows: if a pattern has been trained, then the next presentation of the same pattern must not generate a new winning class. On the contrary, the same F unit should win, with a higher activation than all the other recognition units.

This implies that the norm of the initial weight-vector has to be
smaller than the one it has after several training cycles.
If **J** () is the actual winning unit in F , then
equation is given by the theory:

where ** z** is the the weight vector of the links from the
F units to the **J**th F unit and where d is a parameter, described
below.

If all initial values are presumed to be equal, this means:

If equality is chosen in equation , then ART2 will be as sensitive as possible.

To transform the inequality to an equation, in order to compute values, we introduce another parameter and get:

where .

To initialize an ART2 network, the function ` ART2_Weights` has to
be selected. Specify the parameters d and as the first and
second initialization parameter. (A description of parameter d is given
in the subsection on the ART2 learning function.) Finally press the
-button to initialize the net.

** WARNING!** You should always use ` ART2_Weights` to initialize
ART2 networks. When using another SNNS initialization function the
behavior of the simulator during learning is not predictable, because
not only the trainable links will be initialized, but also the fixed
weights of the network.

Niels.Mache@informatik.uni-stuttgart.de

Tue Nov 28 10:30:44 MET 1995