Stat231_HW1 - Stat 231 CS 276A assignment#1 total 10 points...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Stat 231 - CS 276A assignment #1: total 10 points Due: Oct. 13 Wednesday at class meeting. Policy : You are supposed to finish the homework by yourself. Do not discuss homework with your classmate or friends who took this course before. People who violate this policy, once caught, will be dismissed from the class. You must turn in the homework at class meeting. If you hand in homework in the afternoon, then it will be counted as being late for 1 day. Total 3 late days for the entire class. Questions : My office hour is M 3-4:30pm, BH 9404. Problem 1 : 2 points, (Bayesian classification boundary). For a two-class recognition problem: salmon ( w = 1) and sea bass ( w = 2), suppose we uses two features x = ( x 1 ,x 2 ) (e.g. length and brightness) and the two class models (also called likelihood models) p ( x | w = 1) and p ( x | w = 2) are supposed to be 2D Gaussian distributions centered at points (3 , 6) and (6 , 3) respectively with the same covariance matrix Σ = 3 I ( I is the identity matrix). Suppose the prior probability for salmon and sea bass is p ( w = 1) = 0 . 4 , p ( w = 2) = 0 . 6. 1. Suppose we use a Bayes decision rule, write the two discriminant functions g 1 ( x ) and g 2 ( x ). 2. Derive the equation for the decision boundary
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 11/24/2010 for the course STAT 201a taught by Professor Wu during the Spring '10 term at Pasadena City College.

Page1 / 3

Stat231_HW1 - Stat 231 CS 276A assignment#1 total 10 points...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online