Algebra-2 course for 1st year of a Mathematics

bachelor’s degree.
Chapter 2: Linear Mappings

Dr. Louiza TABHARIT

February 2024

Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 1 / 17

Linear Mappings: Definition
Definition
Let A and B be two vector spaces. A Linear Mapping f : A → B is a
mapping which satisfies the following two properties:
1) Addition linearity: ∀u, v ∈ A : f (u + v ) = f (u ) + f (v ) .
2) Multiplication linearity: ∀α ∈ IF ; ∀u ∈ A : f (αu ) = αf (u ) .
The set of all linear mappings is the vector space denoted by L (A, B ) .

Examples
1 The Identity mapping (id : A → A; id (u ) = u ) is linear.
2 The Zero mapping (O : A → B; O (u ) = 0B ) is linear.
3 The Projection mappings
exmpl : f : R3 → R2 ; f (x, y , z ) = (y , z ) .

4 The mapping: f : R2 → R2 ; f (x, y ) = (x + y , x − y ) is linear.

5 The Homothety with scale factor k (hk : A → A; hk (u ) = ku ) .
Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 2 / 17
Linear Mappings
Theorem (Characterization)
Let A and B be two vector spaces. A mapping f : A → B is Linear iff for
any vectors u, v ∈ A and scalars α, β ∈ IF :

f (αu + βv ) = αf (u ) + βf (v ) .

Example
Let f : R2 → R2 such that f (x, y ) = (x + y , x − y ) . Taking two vectors
u = (a, b ) and v = (a0 , b 0 ) and two scalars α, β ∈ R, we have

f (αu + βv ) = f αa + βa0 , αb + βb 0

αa + βa0 + αb + βb 0 , αa + βa0 − αb + βb 0
       

=
= (αa + αb, αa − αb ) + βa0 + βb 0 , βa0 − βb 0

= αf (a, b ) + βf a0 , b 0 = αf (u ) + βf (v ) .


Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 3 / 17
Linear Form

Definition
A linear form ( or one-form) is a linear mapping from a vector space to
its field of scalars(i.e. f : A → IF ) .
The set of all linear forms is denoted by L (A, IF ) .

Exemple
1 f : R2 → R; f (x, y ) = x + y .
2 g : R3 → R; g (x, y , z ) = 4 (x + 2y − z ) .

Remark
the mapping f : R → R; f (x ) = x + a with a ∈ R∗ is not linear (but is
an affine transformation).

Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 4 / 17

Linear Mappings
Proposition
If the mapping f : A → B is Linear, then:
1 f (0A ) = 0B .
2 ∀u ∈ A : f (−u ) = −f (u ) .

Proof.
for (1) use α = 0 and for (2) use α = −1.

Proposition
Let f : A → B be a Linear mapping and {u1 , · · · , un } a spanning set of A,
we have
n n
∀v ∈ A, ∃ (α1 , · · · , αn ) ∈ IF n : v = ∑ αi ui =⇒ f (v ) = ∑ α i f ( ui )
i =1 i =1

Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 5 / 17

Inverse of linear mapping
Proposition
If f : A → B is a bijective Linear mapping, then its inverse f −1 : B → A is
also linear.

Proof.
Let α, β ∈ IF , and two images a, b ∈ B. Since f is a bijective,
∃!u ∈ A : f (u ) = a, and ∃!v ∈ A : f (v ) = b. So,

f −1 (αa + βb ) = f −1 (αf (u ) + βf (v ))
= f −1 (f (αu ) + f ( βv ))
= f −1 (f (αu + βv ))
= αu + βv
= αf −1 (a) + βf −1 (b ) .

Hence, f −1 is a linear mapping.

Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 6 / 17
Linear mapping Composition
Proposition
If f : A → B and g : B → C are linear mappings, then their composition
g ◦ f : A → C is also linear.

Proof.
Let α, β ∈ IF ; u, v ∈ A :

(g ◦ f ) (αu + βv ) = g (f (αu + βv ))
= g (αf (u ) + βf (v ))
= g (αf (u )) + g ( βf (v ))
= αg (f (u )) + βg (f (v ))
= α (g ◦ f ) (u ) + β (g ◦ f ) (v ) .

Thus, g ◦ f is a linear mapping.

Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 7 / 17
Kernel of a linear mapping
Definition
Let f : A → B be a linear mapping. The Kernel of f is the set :

ker f = {u ∈ A : f (u ) = 0B } = f −1 ({0B }) .

Example
Let f : R2 → R2 ; f (x, y ) = (x + y , x − 3y ) .

u ∈ R 2 : f ( u ) = 0 R2

ker f =
= (x, y ) ∈ R2 : f (x, y ) = (0, 0)


= (x, y ) ∈ R2 : (x + y , x − 3y ) = (0, 0)


= (x, y ) ∈ R2 : x + y = 0 ∧ x − 3y = 0 = {(0, 0)} .



So, the ker f contains only the zero vector ”0R2 ”.

Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 8 / 17
kernel of a linear mapping

Example
Let g : R3 → R2 ; g (x, y , z ) = (x + y + z, −x + 2y + 2z )

u ∈ R3 : f (u ) = 0R2

ker g =
= (x, y , z ) ∈ R3 : f (x, y , z ) = (0, 0)


= (x, y , z ) ∈ R3 : x + y + z = 0 ∧ −x + 2y + 2z = 0


= (x, y , z ) ∈ R3 : y = −z ∧ x = 0


= {(0, −z, z ) /z ∈ R}
= {z (0, −1, 1) /z ∈ R}
ker g = span {a} /a = (0, −1, 1) .

Thus, the ker g contains an infinity of vectors.

Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 9 / 17

kernel of a linear mapping
Proposition
Let f : A → B be a linear mapping.
Prop1: ker f is a subspace of the vector space A.
Prop2: ker f = {0A } ⇔ f is injective.

Proof.
(⇐) Suppose that f is injective. So, we have
f Lin f inj
u ∈ ker f =⇒ f (u ) = 0B = f (0A ) =⇒ u = 0A .
(⇒) Suppose that ker f = {0A } . To prove that f is injective we take
u, v ∈ A, such that

f (u ) = f (v ) =⇒ f (u ) − f (v ) = 0B =⇒ f (u − v ) = 0B
=⇒ u − v ∈ ker f =⇒ u − v = 0A =⇒ u = v .

So, f is injective.
Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 10 / 17
Image of a linear mapping
Definition
Let f : A → B be a linear mapping. The Image of f is the set :

Imf = f (A) = {f (u ) ∈ B : u ∈ A} .

Example
Let f : R2 → R2 ; f (x, y ) = (x + y , x − 3y ) .

Imf = {f (u ) ∈ B : u ∈ A}
(x + y , x − 3y ) ∈ R2 : (x, y ) ∈ R2

=
= {(x, x ) + (y , −3y ) : x, y ∈ R}
= {x (1, 1) + y (1, −3) : x, y ∈ R}
= span {c, d } /c = (1, 1) and d = (1, −3) .

Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 11 / 17

Image of a linear mapping

Example
Let g : R3 → R2 ; g (x, y , z ) = (x + y + z, −x + 2y + 2z ) :

Img = {g (u ) ∈ B : u ∈ A}
(x + y + z, −x + 2y + 2z ) ∈ R2 : (x, y , z ) ∈ R3

=
= {(x, −x ) + (y , 2y ) + (z, 2z ) : x, y , z ∈ R}
= {x (1, −1) + [y + z ] (1, 2) : x, y , z ∈ R}
= span {c, d } /c = (1, −1) and d = (1, 2) .

Exercise: Determine the set Imf such that f : R2 → R2 ;

f (x, y ) = (3x − 6y , x − 2y ) .

Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 12 / 17

Image of a linear mapping

Proposition
Let f : A → B be a linear mapping.
Prop1: Imf is a subspace of the vector space B.
Prop2: Imf = B ⇔ f is surjective.

Proof.
f is surjective ⇔ ∀v ∈ B, ∃u ∈ A : v = f (u ) . Witch means that v ∈ Imf .
Hence, B ⊆ Imf . And, we know that Imf ⊆ B. Then, Imf = B.

Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 13 / 17

The image of a set of vectors

Proposition
Let f : A → B be a linear mapping and let E be a subset of independent
vectors from A and F a spanning set of A.
1 f is injective⇔ The vectors of f (E ) are independent.
2 f is surjective⇔ f (F ) is a spanning set of B.

Corollary
Let f : A → B be a linear mapping and G is a basis of A :

f is bijective ⇔ f (G ) is a basis of B

Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 14 / 17

The image of a set of vectors

Exercise: Let a linear mapping f : R3 → R2 ;

f (x, y , z ) = (x + y , y − z ) Give a basis for the subspace
Imf . Is f surjective?
Solution: we know that G = {(1, 0, 0) ; (0, 1, 0) ; (0, 0, 1)} is a basis of
R3 , then

f (G ) = {f (1, 0, 0) ; f (0, 1, 0) ; f (0, 0, 1)}

= {(1, 0) ; (1, 1) ; (0, −1)} .

It is clear that (1, 1) = (1, 0) − (0, −1) . So,

Imf = span {(1, 0) ; (0, −1
)} . Since Imf ⊆ R2 and
dim (Imf ) = 2 = dim R2 ; we can say that Imf = R2 .
Hence, f is surjective.

Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 15 / 17

The image of a set of vectors

Proposition
Let f : A → B be a linear mapping and let E = {u1 , · · · , un } be a subset
of A.
1 E is L. dependent =⇒ f (E ) is L. dependent.
2 f (E ) is L. independent =⇒ E is L. independent.

Proof.
(Prop1): If E is L. dependent then ∃ (α1 , · · · , αn ) ∈ IF n − {(0, · · · , 0)}
n
such that ∑ αi ui = 0A . So,
 i=
1
n n
f ∑ αi ui = f (0A ) =⇒ ∑ αi f (ui ) = 0B . Thus, f (ui ) are L.
i =1 i =1
dependent.
(Prop2): is the contrapositive of the proposition 1.

Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 16 / 17

Dimension Theorem
Theorem
If f : A → B is a linear mapping, then

dim A = dim (ker f ) + dim (Imf ) .

Corollary
Let f : A → B be a linear mapping, the following holds:
1 f injective =⇒ dim (A) ≤ dim (B ) ;
2 f surjective =⇒ dim (A) ≥ dim (B ) ;
3 f bijective =⇒ dim (A) = dim (B ) .

Remark
The converse implications are not necessarily true.
Dr. Louiza TABHARIT FSEI. Univ-Mosta February 2024 17 / 17