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 Content is under active construction and has not been peer-reviewed.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
/
The EM Algorithm
The EM Algorithm
1 questions
Difficulty 6-6
View topic
Intermediate
0 / 1
1 intermediate
Adapts to your performance
1 / 1
intermediate (6/10)
conceptual
The EM algorithm alternates between computing expected sufficient statistics (E-step) and maximizing the expected complete-data log-likelihood (M-step). Which property is guaranteed?
Hide and think first
A.
EM converges to the global maximum of the log-likelihood
B.
EM converges in a finite number of iterations
C.
The observed data log-likelihood is non-decreasing at every iteration:
ℓ
(
θ
(
t
+
1
)
)
≥
ℓ
(
θ
(
t
)
)
D.
The E-step and M-step each individually decrease the KL divergence between the model and the data
Submit Answer