Scribd
Upload
20 views1 page

HW1

This document contains instructions for a homework assignment involving three questions. The first asks to find eigenvalues of matrices and check properties. The second involves singular value decomposition on movie ratings data to find the k-rank approximation. The third involves principal component analysis on celebrity face images.

Uploaded by

angy_1123
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
20 views1 page

HW1

This document contains instructions for a homework assignment involving three questions. The first asks to find eigenvalues of matrices and check properties. The second involves singular value decomposition on movie ratings data to find the k-rank approximation. The third involves principal component analysis on celebrity face images.

Uploaded by

angy_1123
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
You are on page 1/ 1

Department of Electrical and Computer Engineering

University of Delaware
FSAN/ELEG815 Analytics I: Statistical Learning
Homework #1, Fall 2021

Name:

1. Find the eigenvalues of R and R2 and R−1 :


 
2 1
R=
1 2

Check the trace and the determinant of R, R2 and R−1 using the matrices
and the eigenvalues.
2. Consider the rating of a 1000 movies from 671 viewers given in the reduced
Netflix dataset ‘ratingsData’. Viewers rated movies on a scale from 1
to 5, 0 for movies that were not rated. Using this dataset, repeat the
example of Singular Value Decomposition of the slides. Find the k-rank
approximation for k=4, k=6, k=8 and k=100. For each case, use Mean
Square Error (MSE) to measure how good your predictions are. Use only
the ratings given in the testing dataset ‘ratingsTest’ to compute the MSE.
Compare this error with the initial error i.e. when the testing ratings were
unknown (value of 0).
Note: We attached a notebook that give you hints and also load the Data
for you.
3. Given the dataset of images (Anne Hathaway, Benjamin Bratt, Ryan
Gosling, Keanu Reeves, Meghan Markle):

– Reproduce the results of the example of Principal Component Anal-


ysis showed in slide 74 i.e. the average faces.
– Reproduce the results in slide 76 i.e. show the first 5 eigenvectors
and plot the first 20 eigenvalues.
– Select one of the faces and reproduce the results of slide 77.
– Use a picture of yourself and reproduce the results of slide 81. De-
termine which celebrity you resemble more.
Note: you should only get the first 20 eigenvalues and eigenvectors. You
can resize your pictures as desired depending on your computer resources.
We attached a notebook that give you hints and also load the Dataset for
you.

You might also like