homework3 - CSE 555 Spring 2010 Homework 3 Jason J. Corso...

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

View Full Document Right Arrow Icon
CSE 555 Spring 2010 Homework 3 Jason J. Corso Computer Science and Engineering SUNY at Buffalo jcorso@buffalo.edu Date Assigned 15 Mar 2010 Date Due 5 April 2010 Homework must be submitted in class. No late work will be accepted. This homework contains only written questions. You must turn in the written questions in class. Problem 1: No Free Lunch Theorem (25%) This is problem 1 from chapter 9 DHS. One of the “conservation laws” for generalization states that the positive generalization performance of an algorithm in some learning situations must be offset by the negative performance elsewhere. Consider a very simple learning algorithm that seems to contradict this law. For each test pattern, the prediction of the majority learning algorithm is merely the category most prevalent in the training data. 1. Show that averaged over all two-category problems of a given number of features, the off-training set error is 0.5. 2. Repeat (1) by for the
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.

Page1 / 2

homework3 - CSE 555 Spring 2010 Homework 3 Jason J. Corso...

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