tag:blogger.com,1999:blog-1872817975428307009.post5046812705051269655..comments2009-05-07T20:40:27.801-07:00Comments on cse471/598 Intro to AI Spring 2009 Blog: Thinking Cap qns on Bayes Networks...Subbarao Kambhampatihttp://www.blogger.com/profile/08449853328445416609noreply@blogger.comBlogger3125tag:blogger.com,1999:blog-1872817975428307009.post-33525317973612790192009-03-29T15:30:00.000-07:002009-03-29T15:30:00.000-07:000) 2^m seems correct. But if the variable is alway...0) 2^m seems correct. But if the variable is always true no matter what you put it then isnt it really independent... in which case all the connections to the parents are spurious? Which really begs the question that what ways can we think of to test for conditional independence given a network and cpt's(Say the network was badly designed)?<BR/>One would be just check brute force enumeration.Sidhttps://www.blogger.com/profile/03158438332840577950noreply@blogger.comtag:blogger.com,1999:blog-1872817975428307009.post-50844835755478725692009-03-22T21:50:00.000-07:002009-03-22T21:50:00.000-07:000. The sum should be 2^m, which corresponds to the...0. The sum should be 2^m, which corresponds to the case that this variable is always true regardless of the values of its parents.<BR/><BR/>1. Michael already answered this question.<BR/><BR/>2. The CPT entries should NOT lead to inconsistency, as long as they are specified according to the network topology. The Bayesian network is based on Bayesian perspective of probability, instead of the frequentist point of view. So the subjectiveness is justified in this context.<BR/><BR/>3. No<BR/><BR/>4. I assume the problem statement is “The CIA(B1) is a superset of A and CIA(B2) is a subset of A”. This should NOT be a huge problem, as we can always be more conservative and assume less CIA’s, at the price of assessing more CPT entries. In choosing B1, the resulting network encodes more CIA’s than the domain actually has, while choosing B2 results in network encodes less CIA’s.Shuiwang Jihttps://www.blogger.com/profile/05658112773874389744noreply@blogger.comtag:blogger.com,1999:blog-1872817975428307009.post-15458325876097192652009-03-22T18:22:00.000-07:002009-03-22T18:22:00.000-07:00I'll take a break from finishing project 2 (late) ...I'll take a break from finishing project 2 (late) to give these a try.<BR/><BR/>1. You can check if the JPD from your friend entails the same conditional probabilities from your BN structure. In other words, if the friend's JPD can't be expressed as the product of local distributions than he or she is wrong.<BR/><BR/>2. Not quite understanding this one...sorry.<BR/><BR/>3 and 4 - I've spent too long thinking. I'll come back to these after I finish the project and my other homework.Michaelhttps://www.blogger.com/profile/09847101875727712446noreply@blogger.com