Information Technology Reference
In-Depth Information
Fig. 4.
Probability models for the economic recession event, the messy event, the happy
citizen event and the wrong policy event
An output of a neuron is computed as follows:
o
(
r
)
j
=
f
(
s
(
r
)
j
)
(1)
K
(
r−
1)
s
(
r
)
j
w
(
r
)
j,i
o
(
r
)
+
b
(
r
)
j
=
ʱ
(2)
i
i
1
1+
e
−x
f
(
x
)=
(3)
1)-th layer,
o
(
r−
1)
i
where
K
(
r−
1)
is the number of neurons of the (
r
−
is the
i
-th
1)-th layer,
w
(
r
)
j,i
output of the (
r
−
is the weight from the
i
-thneuronatthe
1)-th layer to the
j
-thneuronatthe(
r
)-th layer,
ʱ
is a scalar and
b
(
r
)
j
(
r
−
is
the
j
-th bias of the
r
-th layer. Note that:
f
(
x
)=
f
(
x
)
∗
(1
−
f
(
x
))
.
(4)
In the third step, we calculate the expectation achievement level of the player
based on the player's current achievement. This is done by computing the differ-
ence between the outputs of the neural network and the four default threshold
Search WWH ::
Custom Search