The network's dynamics are governed by a set of nonlinear coupled differential equations of the form:
or in a general form:
where the corresponds to an input bias to neuron i, and
the
term introduces an element of decay into the network's
dynamics, which can be used to ensure that u variables remain
bounded in magnitude.
To make sure that stability of the of the net is guaranteed, is
updated continuously by evaluating
as a function of
the total input to the
neuron.
where
So, the overall dynamics of the network is governed by equations (7) and (8).