Skip to main content
Theorem
Path
Curriculum
Paths
Labs
Diagnostic
Case Study
Blog
Search
Sign in
Quiz Hub
/
No-Free-Lunch Theorem
No-Free-Lunch Theorem
6 selected
Difficulty 3-6
6 unseen
View topic
Foundation
New
0 answered
3 foundation
3 intermediate
Adapts to your performance
Question 1 of 6
120s
foundation (3/10)
counterexample
Which statement does the No-Free-Lunch theorem *not* support?
Hide and think first
A.
Worst-case distributional bounds may be loose for natural data; many real problems admit good
H
choices and easy generalization.
B.
No algorithm achieves low expected risk on every distribution simultaneously when the hypothesis class is the set of all binary functions.
C.
Choosing a restricted hypothesis class
H
is a form of prior knowledge that NFL forces every successful learner to commit to.
D.
Learning is impossible in general; no algorithm can ever achieve low risk on a useful problem regardless of how much data is available.
Show Hint
Submit Answer
I don't know