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 Content is under active construction and has not been peer-reviewed.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
/
PAC-Bayes Bounds
PAC-Bayes Bounds
3 questions
Difficulty 6-10
View topic
Intermediate
0 / 3
1 intermediate
2 advanced
Adapts to your performance
1 / 3
intermediate (6/10)
conceptual
In the PAC-Bayes bound, you bound the risk of a posterior
Q
over hypotheses as
R
(
Q
)
≤
R
^
(
Q
)
+
2
n
KL
(
Q
∥
P
)
+
l
n
(
n
/
δ
)
. What constraint does the prior
P
satisfy?
Hide and think first
A.
P
must be chosen before seeing the training data (data-independent)
B.
P
must be a Gaussian distribution
C.
P
must be the uniform distribution over
H
D.
P
must assign nonzero probability to the Bayes-optimal hypothesis
Submit Answer