# hw1 - Q 7 (5pts) The textbook describes an algorithm...

This preview shows page 1. Sign up to view the full content.

CSCI 573 2010 Homework Assignment 1 Due: Feb 9th, 2010 before the class 12:30pm (either in person or electronically) 1 Basic probabilities Q properties on page 25, as well as provide a counterexample to the intersection property when the distribution P is positive. Q 2 Bayesian network: representation Q reasonable realistic example. Q Q shows only for a speci±c graph and should not be used as a general proof.) Q properties, we can derive local Markov independencies. Please consult the slide(s) for a hint.
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Q 7 (5pts) The textbook describes an algorithm (Algorithm 3.1 on pp. 75) to determine whether X and Y are d-separated by Z in a Bayesian network. Run this algorithm on the Bayesian network in Fig. 3.6 to determine whether X and W are d-separated by Y . You need to turn in traces of running the algorithm. Namely, show the values of the variables ( L , A for steps 2-9, and L , V and R from steps 12- 34) every time they change. Q 8 (10pts) Exercise 3.28. You do not have to implement in a specic programming language. However, you need to explain your algorithm well, particularly i) why it is correct? ii) why it is more ecient than running Algorithm 3.1 for every possible source variable X i . Specically, what is the computational complexity of your algorithm? 1...
View Full Document

## This note was uploaded on 05/18/2010 for the course CS 573 taught by Professor Sha during the Spring '10 term at USC.

Ask a homework question - tutors are online