Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
3 views

Lecture 07

Uploaded by

n3354418
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
3 views

Lecture 07

Uploaded by

n3354418
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Lecture 07

Eigenvalues and
Eigenvectors
A matrix eigenvalue problem considers the vector equation
Ax = I (1)
Here A is a given square matrix, an unknown scalar, and x an unknown vector. In a matrix
eigenvalue problem, the task is to determine and x’s that satisfy (1). Since is always a solution
for any and thus not interesting, we only admit solutions with x  0.

The solutions to (1) are given the following names: The ’s that satisfy (1) are called
eigenvalues of A and the corresponding nonzero x’s that also satisfy (1) are called eigenvectors
of A.

From this rather innocent looking vector equation flows an amazing amount of relevant theory
and an incredible richness of applications. Indeed, eigenvalue problems come up all the time in
engineering, physics, geometry, numerics, theoretical mathematics, biology, environmental
science, urban planning, economics, psychology, and other areas. In recent times, eigenvalue
problems are playing an important role in machine learning also. Thus, in your career you are
likely to encounter eigenvalue problems. We jump into the problem directly through a problem.

Example 01
Find the eigenvalues and eigenvectors of
3 0 0
5 4 0
3 6 1
To find the eigenvalues, we write
3 0 0 𝑥 𝑥
5 4 0 𝑥 = 𝜆 𝑥
3 6 1 𝑥 𝑥
In terms of matrix, this is written as
Ax – Ix = 0
Or (A – I)x = 0 (2)
As this is a homogeneous equation where the rank and the number of unknowns are same.
Therefore, to find a solution, we force the system to become dependent. To do this, we force
det(A – I)x = 0 (3)
Therefore
3−𝜆 0 0
5 4−𝜆 0 =0
3 6 1−𝜆
Expanding the determinant, we obtain
(3 – )(4 – )(1 – ) = 0
From here, we obtain the eigenvalues to be 1 = 3,  2 = 4, and  3 = 1.

To find the eigenvectors.

1 = 3
Substituting the value of in equation 2, we obtain
5x1 + x2 = 0
3x1 + 6x2 – 2x3 = 0
The fact that we end up with two equations should not be surprising. We have ensured that the
system becomes dependent. Therefore, one of the equations is now dependent on the others.
Solving the system, we obtain the solution x1T = [–2, 10, 27]T.

2 = 4
Substituting the value of in equation 2, we obtain
–x1 = 0
5x1 = 0
3x1 + 6x2 – 3x3 = 0
We obtain the solution to be x1 = 0. Selecting x2 to be , we get x3 = 2. Therefore x T2 = [0, 1, 2]T

Finally,  3 = 1
Substituting the value of in equation 2, we obtain
2x1 = 0
5x1 + 3x2 = 0
3x1 + 6x2 = 0
Solving these, we obtain, x1 = 0, x2 = 0, x3 can be selected arbitrarily, therefore, let x3 = .
Therefore, x 3T = [0, 0, 1]T

Exercise

Find eigenvalues and eigenvectors.


1. 3 0 2. 5 −2
0 −0.6 9 −6
3. 1 2 4. 0 3
2 4 −3 0
5. 1 2 6. 0 1
0 3 0 0
7. 𝑎 𝑏 8. 0.8 0.6
−𝑏 𝑎 −0.6 0.8
9. 𝑐𝑜𝑠 𝑠𝑖𝑛 10. 0 0
−𝑠𝑖𝑛 𝑐𝑜𝑠 0 0
11. 6 2 −2 12. 3 5 3
2 5 0 0 4 6
−2 0 7 0 0 1
13. 13 5 2 14. 2 0 −1
2 7 −8 0 1/2 0
5 4 7 1 0 4

Applications of eigenvalues and eigenvectors

Eigenvalues and eigenvectors have extremely sophisticated applications in many field in science
and engineering. Just some are listed below.

 Structural engineering: Analyzing the stability and vibrational properties of structures


 Quantum mechanics: Representing the possible energy levels of a quantum system and
describing the associated wave functions
 Power systems: Determining the stability and transient response of a system, which helps
with voltage stability analysis and fault detection
 Electromagnetic field analysis: Studying wave propagation, resonant frequencies, and
antenna design
 Data analysis and machine learning: Reducing the dimensionality of data while
preserving the most significant information
 Stability analysis: A classic application in mechanical engineering and architecture
 Google's page rank algorithm: A well-known example of an application

Other applications include:

Vibration analysis, Electric circuits, Advanced dynamics, Atomic orbitals, Facial recognition,
Matrix diagonalization, Image processing, Statistics, Geophysics, and Spectral analysis.

Use of Eigenvalues and Eigenvectors in Machine Learning and AI:


1. Dimensionality Reduction (PCA): In Principal Component Analysis (PCA), you
calculate the eigenvectors and eigenvalues of the covariance matrix of your data. The
eigenvectors (principal components) with the largest eigenvalues capture the most
variance in the data and can be used to reduce the dimensionality of the dataset while
preserving important information.
2. Image Compression: Eigenvectors and eigenvalues are used in techniques like Singular
Value Decomposition (SVD) for image compression. By representing images in terms of
their eigenvectors and eigenvalues, you can reduce storage requirements while retaining
essential image features.
3. Support vector machines: Support vector machines (SVMs) are a type of machine
learning algorithm that can be used for classification and regression tasks. SVMs work by
finding a hyperplane that separates the data into two classes. The eigenvalues and
eigenvectors of the kernel matrix of the SVM can be used to improve the performance of
the algorithm.
4. Graph Theory: Eigenvectors play a role in analyzing networks and graphs. They can be
used to find important nodes or communities in social networks or other interconnected
systems.
5. Natural Language Processing (NLP): In NLP, eigenvectors can help identify the most
relevant terms in a large document-term matrix, enabling techniques like Latent Semantic
Analysis (LSA) for document retrieval and text summarization.
6. Machine Learning Algorithms: Eigenvalues and eigenvectors can be used to analyze the
stability and convergence properties of machine learning algorithms, especially in deep
learning when dealing with weight matrices in neural networks.

You might also like