This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: 10708 Graphical Models: Homework 1 Due October 3rd, beginning of class September 20, 2005 1 [15 pts] Conditional Probability 1.1 [5 pts] Let X , Y , Z be three disjoint sets of variables such that S = X ∪ Y ∪ Z . Prove that P  = ( X ⊥ YZ ) if and only if we can write P in the form: P ( S ) = f ( X , Z ) g ( Y , Z ) 1.2 [3 pt] Is it possible for both f and g above to be probability distributions over their respective sets of variables? Formally, is it possible for every distribution P over ( X ∪ Y ∪ Z ) with the independency above, to be expressed as a product of a distribution over ( X ∪ Z ) and a distribution over ( Y ∪ Z )? Justify your answer. (Hint: Assume the variables are binary; look at the marginal probability of Z ) 1.3 [4 pts] Prove or disprove (by providing a counterexample) each of the following properties of inde pendence: 1. ( X ⊥ Y, W  Z ) implies ( X ⊥ Y  Z ). 2. ( X ⊥ Y  Z ) and ( X, Y ⊥ W  Z ) imply ( X ⊥ W  Z ). 3. ( X ⊥ Y, W  Z ) and ( Y ⊥ W  Z ) imply ( X, W ⊥ Y  Z ). 4. ( X ⊥ Y  Z ) and ( X ⊥ Y  W ) imply ( X ⊥ Y  Z, W ). 1.4 [3 pt] Provide an example of a distribution P ( X 1 , X 2 , X 3 ) where for each i 6 = j , we have that ( X i ⊥ X j ) ∈ I ( P ), but we also have that ( X 1 , X 2 ⊥ X 3 ) / ∈ I ( P ). 1 2 [15 pts] Graph Independencies 2.1 [4 pts] X1 X2 X3 X6 X8 X4 X5 X7 Figure 1: Graphical Model for Prob. 2 Let X = { X 1 , . . . , X n } be a random vector with distribution given by the graphical model in Figure 1. Consider variable X 1 . What is the minimal subset of the variables, A ⊆ X  {...
View
Full
Document
This note was uploaded on 05/25/2008 for the course MACHINE LE 10708 taught by Professor Carlosgustin during the Fall '07 term at Carnegie Mellon.
 Fall '07
 CarlosGustin

Click to edit the document details