Beta. Content is under active construction and has not been peer-reviewed. Report errors on
GitHub
.
Disclaimer
Theorem
Path
Curriculum
Paths
Demos
Diagnostic
Search
Quiz Hub
/
Generative Adversarial Networks
Generative Adversarial Networks
1 questions
Difficulty 6-6
View topic
Intermediate
0 / 1
1 intermediate
Adapts to your performance
1 / 1
intermediate (6/10)
conceptual
In the original GAN formulation, the generator
G
and discriminator
D
play a minimax game:
min
G
max
D
E
[
lo
g
D
(
x
)]
+
E
[
lo
g
(
1
−
D
(
G
(
z
)))]
. At the Nash equilibrium, what is
D
∗
(
x
)
?
Hide and think first
A.
D
∗
(
x
)
=
1/2
for all
x
, since the discriminator cannot distinguish real from generated data
B.
The equilibrium does not exist because the minimax game always oscillates
C.
D
∗
(
x
)
=
p
data
(
x
)
, directly outputting the data density at each input point
D.
D
∗
(
x
)
=
1
for real data and
D
∗
(
x
)
=
0
for generated data
Submit Answer