2
$\begingroup$

Say one is given a probabilistic graphical model and a cut of the underlying graph. Do we know any statements about when and how can one or many of the marginals (of the sources) or the conditionals (along the edges) in one half be altered such that no measurement on the other-side will know about it?

Improve this question
$\endgroup$
1
  • $\begingroup$ This question might get better answers on stats.stackexchange.com. $\endgroup$
    – Qiaochu Yuan
    Commented Apr 8, 2016 at 16:48

0

Reset to default

You must log in to answer this question.