Bayesian networks Flashcards
1
Q
Why do we want to use Bayesian Networks?
A
To compactly represent probability distributions (belief /likelihood of observations)
2
Q
Does a fully-connected Bayesian network provide any advantage over a full joint probability table representation?
A
No, they would be the same.
3
Q
What is the Markov blanket of a Bayesian network node “n”?
A
It is the set of nodes that makes “n” conditionally independent of all the others. This set of nodes is made of: “n”s parents, children and children’s parents.
