Biomedical Engineering Reference
In-Depth Information
Fig. 1.
Schematic of a Neuron
intractable, models such as the Fitzhugh-Nagumo model have been cre-
ated as qualitative models of dendritic-somatic-axonal interactions
11
. An
artificial neuron is a minimalist qualitative model that suggests that
a neuron integrates variations in potential over time and res if a thresh-
hold is exceeded. An Artificial Neural Network (ANN) is a network of
interconnected articial neurons.
Specically, in an ANN, axonal action potentials are represented by
activations which are considered to be in [0; 1], and synaptic activity is
considered to be governed by a parameter known as a synaptic weight. If x
j
denotes the activation from the j
th
neuron in a network, and if w
ij
is the
synaptic weight of the connection between the i
th
and j
th
articial neuron,
then the activation of the i
th
neuron is
0
@
X
j6=i
1
A
x
i
=
w
ij
x
j
i
(1)
where the activation function () is of the form
1
1 + e
t
(t) =
where > 0 is a parameter
12
.
Fig. 2.
An Articial Neuron
Search WWH ::
Custom Search