Matrix Analysis: Eigen & Rank
Module 1 – Artificial Intelligence &
Machine Learning
Introduction to Matrices
• - A matrix is a rectangular array of numbers
arranged in rows and columns.
• - Used to represent data, perform
transformations, and solve systems of
equations.
• - Notation: A = [aᵢⱼ], where i = row, j = column.
Rank of a Matrix
• - Rank = maximum number of linearly
independent row or column vectors.
• - Full Rank: rank = min(no. of rows, no. of
columns).
• - Rank tells us about the dimension of the
column space.
Eigenvalues and Eigenvectors
• - For a square matrix A, if Av = λv, then:
• v is an eigenvector, λ is its eigenvalue.
• - Represents directions where transformation
stretches/compresses.
• - Found by solving: det(A - λI) = 0
Matrix Diagonalization
• - A matrix A is diagonalizable if A = PDP⁻¹,
where:
• P = eigenvector matrix, D = diagonal matrix of
eigenvalues.
• - Useful in simplifying matrix powers and
understanding matrix structure.
Application in ML: Principal
Component Analysis (PCA)
• - PCA uses eigenvectors of the covariance
matrix.
• - Finds directions of maximum variance
(principal components).
• - Used for dimensionality reduction.
Example: Find Eigenvalues
• Given A = [[4, 2], [1, 3]], find the eigenvalues.
• Step 1: det(A - λI) = 0
• Step 2: |4-λ 2|
• |1 3-λ| = 0
• Step 3: Solve quadratic equation: λ² - 7λ + 10 =
0
Class Activity
• - Find the eigenvalues and eigenvectors of A =
[[2, 1], [1, 2]].
• - Determine the rank of A = [[1, 2, 3], [2, 4, 6],
[3, 6, 9]].
• - Discuss the significance of zero eigenvalues.