0
$\begingroup$

Hello this is most definitely a question for dummies i feel. The question concern bayesian network and inference thereof. I've heard about the product rule, bayes theorem and the chain rule. However this bayesian network still keeps eluding me... So i hope someone could give a quick recap on how i should think the following assignment which i have created based on some research.

So giving the following problem statement:

The Bayesian Network LMV below has three nodes for boolean variables, L, M and V. The probabilities for L and M are: P (M = true) = 0.2 and P (L = true) = 0.7. The conditional probabilities for variable V are as shown in the table below:

Image of table and network, no image seeing as not enough rep...

Question: What value is the value of P(V = false | L = false) ?

How is this computed? This realy confuses me because i dont get how we can isolate P(V = False | L= False) when the image suggest that we only have the conditional probability of V given both L and M? Any help appreciated :)

$\endgroup$
0
$\begingroup$

The conditional probability is : $$\begin{align}\mathsf P(V{=}f\mid L{=}f)~&=~ \dfrac{\mathsf P(V{=}f, L{=}f)}{\mathsf P(L{=}f)}&&\text{by definition}\\[3ex]&=~\dfrac{\displaystyle\sum\limits_{m\in\{f,t\}} \mathsf P(V{=}f,L{=}f,M{=}m)}{\mathsf P(L{=}f)}&&\text{by Law of Total Probability}\\[3ex] &=~\dfrac{\displaystyle\sum\limits_{m\in\{f,t\}}\mathsf P(V{=}f\mid L{=}f,M{=}m)~\mathsf P(L{=}f)~\mathsf P(M{=}m)}{\mathsf P(L{=}f)}&&\text{by DAG Factorisation}\\[1ex] &~~\vdots\end{align}$$

Well, you can take it from here, I am sure.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.