How does one compute this bayesian probability?

30 Views Asked by At

Assume you have a network as follow (where X->Y implies X is the parent of Y)

A->D, B->D

How does one compute $P(A,B|D)$?

A and B are independent so my intuition tells me

$P(A,B|D)= P(A|D)\times P(B|D) = [\frac{P(D|A)}{P(A)}]\times[\frac{P(D|B)}{P(B)}]$

but I am not sure.

1

There are 1 best solutions below

0
On BEST ANSWER

No.   You cannot infer conditional independence from pairwise independence.

Applying Bayes' Rule can only obtain:

$$\begin{align} \mathsf P(A, B\mid D) ~=~& \dfrac{\mathsf P(D\mid A, B)~\mathsf P(A,B)}{\mathsf P(D)} \\[1ex] =~& \dfrac{\mathsf P(D\mid A, B)~\mathsf P(A)~\mathsf P(B)}{\mathsf P(D)} \end{align}$$