I am working on image processing and my probability theory knowledge is low. My question here is I am working with 2 variables X and Y which is dependent on each other. That is we can compute P(X|Y) or P(Y|X). The distribution of each is normal and X not equal to Y.

Baysian Network is more a directed relationship modeling, i.e only one what. If I did understand correctly than I can compute P(X|Y) but not P(Y|X). Is this correct?

My problem is as I stated above i want to model 2 variables that are dependent on each other. I read through Markov Random Fields but I am not sure if it can represent what I need. Please advice me of any possible representation (distribution) I can use to model the 2 variable. Thank you


1 Answer 1


Since both variables are Gaussian, the natural model for their joint distribution is a multivariate Gaussian distribution.


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.