Scribd
Upload
175 views1 page

CSE 455/555 Spring 2013 Quiz Solutions

This document contains solutions to two problems from a quiz for a machine learning course. Problem 1 asks for the definition of a decision boundary in one sentence. The solution provided is that a decision boundary is the region of a problem space in which the output label of a classifier is ambiguous. Problem 2 asks for the inverse of a 2x2 matrix. The solution shows the work using Gaussian elimination to arrive at the inverse, which is given as [-1/4 1; 1 -2].

Uploaded by

Alireza Zahra
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
175 views1 page

CSE 455/555 Spring 2013 Quiz Solutions

This document contains solutions to two problems from a quiz for a machine learning course. Problem 1 asks for the definition of a decision boundary in one sentence. The solution provided is that a decision boundary is the region of a problem space in which the output label of a classifier is ambiguous. Problem 2 asks for the inverse of a 2x2 matrix. The solution shows the work using Gaussian elimination to arrive at the inverse, which is given as [-1/4 1; 1 -2].

Uploaded by

Alireza Zahra
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Prof. Corso, jcorso@buffalo.

edu, SUNY Buffalo


Prepared by David Johnson

CSE 455/555 Spring 2013 Quiz 1 of 14

Solutions
Problem 1: Recall (2pts) (Answer in one sentence only.)
What is a decision boundary?
A decision boundary is the region of a problem space in which the output label of a classifier is ambiguous.
Half-credit was given for answers that captured the general idea of a decision boundary (i.e. it is the division between
two classifier regions), but were not quite correct. Common mistakes included:
The implication that the boundary must be linear (e.g. a point, line, plane or hyperplane)
The implication that the boundary is an inherent property of the classification problem itself rather than the learned
classifier (e.g. that it is the division between two classes)
Overly long and detailed answers (most of which would have lost credit for the one of the above mistakes anyway)the
question says answer in one sentence only and all of these recall questions will be designed to be easily answered
in one sentence. If you feel like you need more, you probably dont have entirely the right answer. Note: this includes
answers that made use of visual aids.

Problem 2: Work (8 pts) (Show all derivations/work and explain.)




4 4
?
What is the inverse of
1 2
 1

1
2
Answer:
41 1
This can be solved via the block method, gaussian elimination or the trick for computing 2 by 2 matrix inverses (swap diagonal, negate off-diagonal, divide each element by determinant). An example using the gaussian elimination method:


4 4 1
1 2 0

2 0
=
1 2

1 0
=
1 2

1 0
=
0 2

1 0
=
0 1


0
: R1 2 R2
1

1 2
: R1 12
0 1

1
2 1
: R2 R1
0 1

1
1
2
: R2 21
1
2 2

1
1
2
14 1

The matrix remaining on the right after converting the matrix on the left to I is the inverse.

You might also like