,

LECTURE NOTES ON ALGEBRA

(MI1141E)

TUAN ANH DAO

SCHOOL OF APPLIED MATHEMATICS AND INFORMATICS

HANOI UNIVERSITY OF SCIENCE AND TECHNOLOGY

Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST

 ,

Chapter 4: Linear mappings

1 BASIC CONCEPTS AND PROPERTIES

2 MATRIX OF A LINEAR MAPPING

3 EIGENVALUES AND EIGENVECTORS

Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST

 ,

Chapter 4: Linear mappings

Lesson 1:
BASIC CONCEPTS AND PROPERTIES

Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST

 ,

Lesson 1: Basic concepts and properties

I. Basic concepts
1. Definition: Let X , Y be two vector spaces on a field K (K = R or
K = C). A mapping f : X → Y
x 7→ f (x)
is called a linear mapping if the following two conditions are satisfied:
• Addition (+): f (u + v ) = f (u) + f (v ) for any u, v ∈ X .
• Scalar multiplication (·): f (k.u) = k.f (u) for any k ∈ K, u ∈ X .
Examples:
a) f : R 3 → R2
(x1 , x2 , x3 ) 7→ f (x1 , x2 , x3 ) = (x1 + 2x2 , x2 − x3 )
b) f : V → V
x 7→ f (x) = x. We call it the identity mapping (IdV ).
2. Remarks:
• If X = Y (= V ), then a linear mapping f : V → V is called a
linear transformation (or linear operator).
• The above two conditions are equivalent to
f (k.u + ℓ.v ) = k.f (u) + ℓ.f (v ) for any k, ℓ ∈ K and u, v ∈ X .
Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST
 ,

Lesson 1: Basic concepts and properties

3. Basic properties:
• If f : X → Y and g : X → Y are two linear mappings. Then,
h =f +g :X →Y
x 7→ h(x) = f (x) + g (x) is also a linear mapping.
• If f : X → Y be a linear mapping and k ∈ K. Then,
h = k.f : X → Y
x 7→ h(x) = k.f (x) is also a linear mapping.
• If f : X → Y and g : Y → Z are two linear mappings. Then,
h =g ◦f :X →Z
x 7→ h(x) = g (f (x)) is also a linear mapping.
II. Image, Kernel and Rank of linear mappings
1. Image and Kernel: Let f : X → Y be a linear mapping. Then,
• The image of f , denoted by Im(f ), is defined by
Im(f ) = f (X ) = {f (x) : x ∈ X },
• The kernel of f , denoted by Ker (f ), is defined by
Ker (f ) = f −1 ({θY }) = {x ∈ X : f (x) = θY } with θY = zero vector.
Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST
 ,

Lesson 1: Basic concepts and properties

2. Theorem:
• Ker (f ) and Im(f ) are subspaces of X and Y , respectively.
• Relation among dimensions: dim(Ker (f ))+ dim(Im(f ))= dim(X ).
3. Rank of a linear mapping: Let f : X → Y be a linear mapping. Then,
rank of f is denoted by rank(f ) and is defined by
rank(f ) = dim(Im(f )).
Examples: Let f : R3 → R2
(x1 , x2 , x3 ) 7→ f (x1 , x2 , x3 ) = (x1 + 2x2 , x2 − x3 )
be a linear mapping.
a) Determine a basis and the dimension of Ker (f ) and Im(f ).
b) Find rank(f ).
4. Some properties: Let f : X → Y be a linear mapping.
• X = span{v1 , v2 , · · · , vn }, then Im(f ) = span{f (v1 ), f (v2 ), · · · , f (vn )}.
• f is injective if and only if Ker (f ) = {θX }.
• Pay attention: If f is bijective, then we call f an isomorphism, and we
say that X and Y are isomorphic, denoted by X ∼ = Y.
Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST
 ,

Chapter 4: Linear mappings

Lesson 2:
MATRIX OF A LINEAR MAPPING

Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST

 ,

Lesson 2: Matrix of a linear mapping

I. Matrix of a linear mapping with respect to given bases
1. Main problem: Let f : X → Y
x 7→ f (x) be a linear mapping.
Assume that S1 and S2 are given bases of X and Y , respectively.

Does there exist a fixed matrix A such that [f (u)]S2 = A.[u]S1 ?

Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST
 ,

Lesson 2: Matrix of a linear mapping

2. Definition: A is called the matrix of f if the following relation holds:

[f (u)]S2 = A.[u]S1 for all u ∈ X .

3. How to find the matrix A:

Assume that S1 = {u1 , u2 , · · · , un } and S2 = {v1 , v2 , · · · , vm }
• Step 1: Replace u by u1 , u2 , · · · , un to compute

[f (u1 )]S2 , [f (u2 )]S2 , · · · , [f (un )]S2 .

• Step 2: A is the matrix containing all coordinates derived from Step 1.

Examples: Let f : R3 → R2
(x1 , x2 , x3 ) 7→ f (x1 , x2 , x3 ) = (x1 + 2x2 , x2 − x3 )
be a linear mapping. Determine the matrix of f with respect to the
following two bases:
a) E1 and E2 are standard bases of R3 and R2 , respectively.
b) S1 = {u1 = (1, 0, 0), u2 = (1, 1, 2), u3 = (1, 2, 3)} is a basis of R3
and S2 = {v1 = (1, 0), v2 = (1, 1)} is a basis of R2 .
Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST
 ,

Lesson 2: Matrix of a linear mapping

4. Some properties:
• If f : X → Y and g : X → Y are two linear mappings. Assume that S1
and S2 are two give bases of X and Y . Moreover, A and B are the
matrices of f and g , respectively, with respect to S1 and S2 . Then,
▶ A + B is the matrix of
h =f +g :X →Y
x 7→ h(x) = f (x) + g (x) with respect to S1 and S2 .
▶ k.A is the matrix of
h = k.f : X → Y
x 7→ h(x) = k.f (x) with respect to S1 and S2 .
• If f : X → Y and g : Y → Z are two linear mappings. Assume that S1 ,
S2 and S3 are two give bases of X , Y and Z . Moreover, A is the
matrix of f with respect to S1 and S2 , and B is the matrix of g with
respect to S2 and S3 . Then, B.A is the matrix of
h =g ◦f :X →Z
x 7→ h(x) = g (f (x)) with respect to S1 and S3 .

Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST

 ,

Lesson 2: Matrix of a linear mapping

II. Matrix of a linear transformation with respect to a given basis
1. Definition 1: Let f : V → V
x 7→ f (x) be a linear transformation.
Assume that S is a given basis of V . Then, A is called the matrix of
f with respect to S if
[f (u)]S = A.[u]S for all u ∈ V .
Examples: Let f : R2 → R2
(x1 , x2 ) 7→ f (x1 , x2 ) = (6x1 + 4x2 , −3x1 − x2 )
be a linear transformation. Determine the matrix of f with respect
to the basis S = {u1 = (1, −1), u2 = (2, 1)}
2. Theorem: (Change of bases)
Let f : V → V be a linear transformation, S1 and S2 be two given
bases of V . Assume that A1 and A2 is the matrix of f with respect
to S1 and S2 , respectively. Then,
A2 = C −1 .A1 .C ,
where C stands for the change-of-basis matrix from S1 to S2 .
Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST
 ,

Lesson 2: Matrix of a linear mapping

3. Definition 2: (Similarities) We call that two n-square matrices A

and B are similar if there exists an invertible matrix P such that

A = P.B.P −1 ⇔ B = P −1 .A.P

Then, we denote A ∼ B or B ∼ A.
• Remarks:
▶ Two matrices A1 and A2 from Theorem (Change of bases)
are similar.
▶ If A and B are two similar matrices, then detA=detB.
6 4
• Question: Let A = . Find a matrix B fulfilling B ∼ A.
−3 −1

Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST

 ,

Chapter 3: Linear mappings

Lesson 3:
EIGENVALUES AND EIGENVECTORS

Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST

 ,

Lesson 3: Eigenvalues and Eigenvectors

I. Eigenvalues and Eigenvectors of a matrix
1. Definition: Let A be a n-square matrix. Then,
• A real number λ is called an eigenvalue of A if there exists a vector
x = (x1 , x2 , · · · , xn ) ̸= (0, 0, · · · , 0) such that
   
x1 x1
 x2   x2 
A.  .  = λ.  .  (∗)
   
 ..   .. 
xn xn

• A vector x = (x1 , x2 , · · · , xn ) ̸= (0, 0, · · · , 0) enjoying (∗) is called

an eigenvector of Acorresponding  to the eigenvalue λ.
2 −3
Examples: Let A = . Then, λ = −2 and u = (3, 4) are an
4 −5
eigenvalue and an eigenvector of A, respectively.
2. Remark: If x is an eigenvector of A, then k.x for any k ∈ R is also
an eigenvector of A. Hence, there are infinitely many eigenvectors!
Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST
 ,

Lesson 3: Eigenvalues and Eigenvectors

3. How to compute eigenvalues and eigenvectors of A:

   
x1 0
x2  0
• Eigenvalues: (∗) ⇔ (A − λ.I).  .  =  . .
   
 ..   .. 
xn 0
This system has at least one solution
x = (x1 , x2 , · · · , xn ) ̸= (0, 0, · · · , 0) ⇔ det(A − λ.I) = 0.
Remark: We call det(A − λ.I) = 0 the characteristic equation of A.
• Eigenvectors: Solve the above system to find solutions
x = (x1 , x2 , · · · , xn ) ̸= (0, 0, · · · , 0) corresponding to the obtained
eigenvalues λ.
   
2 −3 0 0
Examples: Let A = and B = . Find eigenvalues
4 −5 0 0
and eigenvectors of A and B.

Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST

 ,

Lesson 3: Eigenvalues and Eigenvectors

II. Eigenvalues and Eigenvectors of a linear transformation
1. Definition: Let f : V → V
x 7→ f (x) be a linear transformation. Then,
• A real number λ is called an eigenvalue of f if there exists a vector
u ̸= θV such that
f (u) = λ.u (∗∗)
• A vector u ̸= θV enjoying (∗∗) is called an eigenvector of f
corresponding to the eigenvalue λ.
Examples: Let f : R2 → R2
(x1 , x2 ) 7→ f (x1 , x2 ) = (3x1 + x2 , x1 + 3x2 )
be a linear transformation. Then, λ = 2 and u = (1, −1) are an
eigenvalue and an eigenvector of f , respectively.
2. Properties: Let f : V → V be a linear transformation.
• λ is an eigenvalue of f ⇔ h = f − λ.IdV is not injective.
• Assume that λ1 , λ2 , · · · , λn are eigenvalues of f and v1 , v2 , · · · , vn
are eigenvectors corresponding to λ1 , λ2 , · · · , λn . Then, the set
{v1 , v2 , · · · , vn } are linearly independent.
Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST
 ,

Lesson 3: Eigenvalues and Eigenvectors

3. How to compute eigenvalues and eigenvectors of f :
• Step 1: Find the matrix A of f with respect to an arbitrary basis S of
V . As usual, S should be the standard basis E .
• Step 2: Solve the characteristic equation det(A − λ.I) = 0 to find
eigenvalues of A, which are just eigenvalues
  off .
x1 0
x2  0
• Step 3: Solve the system (A − λ.I).  .  =  .  to find solutions
   
 ..   .. 
xn 0
x = (x1 , x2 , · · · , xn ) ̸= (0, 0, · · · , 0) corresponding totheobtained
x1
 x2 
eigenvalues λ. Hence, an eigenvector of f is [u]S =  .  ⇒ u =?
 
 .. 
xn
Examples: Let f : R2 → R2 with f (x1 , x2 ) = (6x1 + 4x2 , −3x1 − x2 ) be
a linear transformation. Find eigenvalues and eigenvectors of f .
Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST
 ,

Lesson 3: Eigenvalues and Eigenvectors

III. Diagonalizations
Main question: For a given matrix A, whether there is a diagonal matrix
A′ such that A′ ∼ A or not?
1. Definition: A is called a diagonalizable matrix if there is a diagonal
matrix A′ such that A′ ∼ A, i.e. there exists an invertible matrix P
such that
A = P.A′ .P −1 ⇔ A′ = P −1 .A.P
• Finding the matrix P is called the diagonalization process of A.
• P is called the matrix diagonalizing A.
     
5 2 −2 1 4 0
Examples: Let A = . Then, P = and A′ = .
2 8 1 2 0 9
2. Theorem: (Conditions for a diagonalizable matrix)
Let A be a n-square matrix. Then,
• A is diagonalizable ⇔ A has n linearly independent eigenvectors.
• A has n distinct eigenvectors ⇒ A is diagonalizable.
Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST
 ,

Lesson 3: Eigenvalues and Eigenvectors

3. How to diagonalize a matrix A:
• Step 1: Solve the characteristic equation det(A − λ.I) = 0 to find
eigenvalues λ1 , λ2 , · · · , λn of A.
• Step 2: For each λk , let us solve the system (A − λk .I).X = O to find
eigenvectors vk = (x1 , x2 , · · · , xn ) ̸= (0, 0, · · · , 0).
• Step 3: Check it out again that the condition for a diagonalizable
matrix is satisfied or not. If yes, then we conclude that
 
λ1 0 0 · · · 0
 0 λ2 0 · · · 0 
P = v1 , v2 , · · · , vn and A′ = 
 
,
· · · 
0 0 0 · · · λn

where the matrix P is written in the

 form ofcolumn vector.
3 1 1
Examples: Let us diagonalize A = 1 3 1.
1 1 3
Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST
 ,

The end

Thank you for your attention!

Tuan Anh Dao, SAMI-HUST ALGEBRA I ♡ HUST