This preview shows pages 1–2. 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: Statistical Pattern Recognition University of Maryland, College Park September 24, 2009 Professor Rama Chellappa Handout 3 Homework 3 This problem set is due Thursday October 1 at 11:00AM . 1. Consider a onedimensional twocategory classification problem with equal priors, P ( 1 ) = P ( 2 ) = 1 / 2, where the densities have the form p ( x  i ) = braceleftbigg x < i e i x x where the i for i = 1 , 2, are positive but unknown parameters. (a) Confirm that the distributions are normalized. (b) The following data were collected: D 1 = { 1 , 5 } and D 2 = { 3 , 9 } for 1 and 2 respectively. Find the maximumlikelihood values 1 and 2 . (c) Given your answer to part (b), determine the decision boundary x * for minimum classification error. Be sure to state which category has higher values than x * , and which category lower values than x * . (d) What is the expected error of your classifier in part (c)?...
View Full
Document
 Fall '05
 staff

Click to edit the document details