Skip to main content
Theorem
Path
Curriculum
Paths
Labs
Diagnostic
Blog
Search
Sign in
Quiz Hub
/
K-Nearest Neighbors
K-Nearest Neighbors
2 selected
Difficulty 2-3
2 unseen
View topic
Foundation
New
0 answered
2 foundation
Adapts to your performance
Question 1 of 2
120s
foundation (2/10)
compare
A learning method is called PARAMETRIC if which property holds?
Hide and think first
A.
The model has a FIXED-SIZE parameter vector
θ
∈
R
d
that does not grow with the training set size
n
.
B.
The model assumes a Gaussian likelihood; non-parametric methods make no distributional assumptions about the data.
C.
The training procedure has a closed-form solution rather than requiring iterative optimization (gradient descent, EM, etc.).
D.
The model has any hyperparameters at all (learning rate,
k
in KNN, depth in trees) that the user picks before training.
Show Hint
Submit Answer
I don't know