0% found this document useful (0 votes)
34 views11 pages

Linear Applications

This document discusses linear applications between vector spaces, defining key concepts such as linearity, image, kernel, rank, and various types of mappings including endomorphisms, isomorphisms, and automorphisms. It provides definitions, propositions, and examples to illustrate these concepts, including the Rank Theorem which relates the dimensions of the kernel and image of a linear application. Additionally, it includes solved exercises to reinforce understanding of the material.

Uploaded by

usemidea
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
0% found this document useful (0 votes)
34 views11 pages

Linear Applications

This document discusses linear applications between vector spaces, defining key concepts such as linearity, image, kernel, rank, and various types of mappings including endomorphisms, isomorphisms, and automorphisms. It provides definitions, propositions, and examples to illustrate these concepts, including the Rank Theorem which relates the dimensions of the kernel and image of a linear application. Additionally, it includes solved exercises to reinforce understanding of the material.

Uploaded by

usemidea
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/ 11

University of Setif I

Course of Algebra II
Pr. Abdelbaki Merouani
Chapter 2

LINEAR APPLICATIONS

2.1 Linear Application


Definition 14. Given (E, +, ·) and (F, +, ·) as two K-vector spaces, an application f from
E to F is called a linear application if and only if:
1. For all u, v ∈ E: f (u + v) = f (u) + f (v)
2. For all u ∈ E and for all α ∈ K: f (αu) = αf (u)
The set of linear applications from a vector space E to a vector space F is denoted by
L(E, F ).
remark 6 (Equivalent Definition). Given (E, +, ·) and (F, +, ·) as two K-vector spaces, an
application f from E to F is a linear application if and only if:
∀u, v ∈ E; ∀α, β ∈ K : f (αu + βv) = αf (u) + βf (v)
remark 7. In the definition, ’+’ and ’·’ represent the composition laws on E and F . It
is important to note that the sets E and F may contain a priori different objects, and the
operations on these different objects may not have a priori connection. It is useful to consider
which law is being referred to when reading the above definition.
Proposition 7. Let f be a linear application from E to F .
1. f (0E ) = 0F
2. For all u ∈ E: f (−u) = −f (u)
Proof. 1. f (0E ) = f (0E + 0E ) = f (0E ) + f (0E ) =⇒ f (0E ) = 0F
2. f (−u) + f (u) = f (−u + u) = f (0E ) = 0F =⇒ f (−u) = −f (u)

Example 13. 1. The application f : R2 → R defined by u = (x, y) 7→ f (u) = 2x − y is a


linear application. Indeed,
(a) For all u = (x, y) and v = (x0 , y0 ) ∈ R2 :
f (u + v) = f ((x, y) + (x0 , y0 ))
= f (x + x0 , y + y0 )
= 2(x + x0 ) − (y + y0 )
= (2x − y) + (2x0 − y0 )
= f (u) + f (v)

16
(b) For all u = (x, y) ∈ R2 and for all λ ∈ K:

f (λu) = f (λ(x, y))


= f (λx, λy)
= 2(λx) − (λy)
= λ(2x − y)
= λf (u)

2. The application f : R2 → R defined by u = (x, y) 7→ f (u) = 2x − 3 is not a linear


application. Indeed, f (0R2 ) ̸= 0R .
remark 8. 1. The sum of two linear applications is a linear application.
2. The product of a linear application by a scalar is a linear application.
3. The composition of two linear applications is a linear application.

2.2 Image and Kernel of a Linear Application


2.2.1 Image of a Linear Application
Definition 15. Given f : E → F as a linear application, the image of f is the subset of F
consisting of the images of elements of E under f . It is denoted by Im(f ) = {f (u) | u ∈ E}.
In other words, v ∈ Im(f ) if and only if there exists u ∈ E such that v = f (u).
remark 9. A linear application f : E → F is surjective if and only if Im(f ) = F .
Proposition 8. The image of f (Im(f )) is a subspace of the vector space F .
Proof. 1. 0F ∈ Im(f ) because f (0E ) = 0F .
2. If v1 , v2 ∈ Im(f ), then for all α1 , α2 ∈ K, we have α1 v1 + α2 v2 = f (α1 u1 + α2 u2 ) for
certain u1 , u2 ∈ E, so α1 v1 + α2 v2 ∈ Im(f ).

Example 14. Consider the linear application f : R3 → R3 defined by u = (x, y, z) 7→ f (u) =


(2y + z, 2x + z, x − y). We will:
1. Determine its image.
2. Check if it is surjective.
Solution
1. Im(f ) = {(x, z, x) | x, z ∈ R}.

Imf = {f (u); u ∈ R3 }
= {f (x, y, z); (x, y, z) ∈ R3 }
= {(2y + z, 2x + z, x − y); (x, y, z) ∈ R3 }
= {(0, 2x, x) + (2y, 0, −y) + (z, z, 0); (x, y, z) ∈ R3 }
= {(0, 2, 1)x + y(2, 0, −1) + z(1, 1, 0); (x, y, z) ∈ R3 }

2. It is easy to verify that v1 (0, 2, 1), v2 (2, 0, −1), v3 (1, 1, 0) are linearly dependent, so
dim(Imf ) ≤ 3, which proves that Imf ̸= R3 , and f is not surjective.

17
2.2.2 Kernel of a Linear Application
Definition 16. Given f : E → F as a linear application, the kernel of f is the subset of E
consisting of the preimages of the zero vector in the space F .
It is denoted by ker(f ) = {u ∈ E | f (u) = 0F }.
In other words, u ∈ ker(f ) if and only if f (u) = 0F .

remark 10. A linear application f : E → F is injective if and only if ker(f ) = {0E }.


Indeed, suppose f is injective. If u ∈ ker f , then f (u) = 0F = f (0E ), so u = 0E .
Conversely, assume that ker f = {0E }. If f (u1 ) = f (u2 ), then we have

f (u1 − u2 ) = f (u1 ) − f (u2 ) = 0F .

So, u1 − u2 ∈ ker f , meaning u1 = u2 .

Proposition 9. The kernel of f (ker(f )) is a subspace of E.

Proof. 1. 0E ∈ ker(f ) because f (0E ) = 0F .

2. If u1 , u2 ∈ ker(f ), then for all α1 , α2 ∈ K:

f (α1 u1 + α2 u2 ) = α1 f (u1 ) + α2 f (u2 )


= α1 · 0F + α2 · 0F
= 0F ,

so α1 u1 + α2 u2 ∈ ker(f ).

Example 15. Consider the linear application f : R3 → R3 defined by

f (x, y, z) = (2y + z, 2x + z, x − y).

1. Determine its kernel.

2. Is the application f injective?

Solution.

1. We have (x, y, z) ∈ ker f if and only if f (x, y, z) = (0, 0, 0), i.e.,

2y +z = 0 (1)




2x + z = 0 (2)
= 0 (3)


x−y

2. From the third equation, we get x = y. Substituting into equation (2), we obtain

2y + z = 0, which implies z = −2y.

Thus, ker f = {(y, y, −2y) | y ∈ R} =


̸ {(0, 0, 0)}.
The kernel of f is not reduced to the zero vector, and f is not injective.

18
2.3 Rank of an Application. Rank Theorem
Definition 17. Let E and F be two finite-dimensional K-vector spaces, and f ∈ L(E, F ).
The rank of f , denoted rg(f ), is the natural number defined by:

rg(f ) = dim(Im(f ))

Theorem 2.3.1 (Rank Theorem). Every linear application f : E → F from a finite-


dimensional vector space E satisfies:

dim(E) = dim(ker(f )) + rg(f ) = dim(E) = dim(ker(f )) + dim(Im(f ))

Example 16. Consider the linear application f defined by

f : R2 [X] → R2

P 7→ f (P ) = (P (0), P (2))
We want to determine the rank of f .
Using the Rank Theorem,

ker f = {P ∈ R2 [X] | f (P ) = 0}
ker f = {P ∈ R2 [X] | (P (0), P (2)) = (0, 0)}
ker f = {P = aX 2 + bX + c | (c, 4a + 2b + c) = (0, 0)}
ker f = {P = aX 2 + bX + c | c = 0, 4a + 2b + c = 0}
ker f = {P = aX 2 + bX | c = 0, b = −2a}
ker f = {aX 2 − 2aX | a ∈ R}
ker f = Span{X 2 − 2X}
dim(ker f ) = 1

According to the Rank Theorem,

dim(R2 [X]) = dim(ker f ) + rg f

rg f = 2

2.4 Endomorphism, Isomorphism, Automorphism


Definition 18 (Endomorphism). A linear map from a vector space E to itself is called an
endomorphism. The set of endomorphisms of E is denoted L(E) or Endo(E).

Example 17. The map


f : R2 → R2
u = (x, y) 7→ f (u) = (2x − y, x + y)
is an endomorphism.

Definition 19 (Isomorphism). A linear map from a vector space E to a vector space F


that is bijective is called an isomorphism. The set of isomorphisms from E to F is denoted
Iso(E, F ).

19
Example 18. 1. The map
f : R2 → R2
u = (x, y) 7→ f (u) = (x + y, x − y)
is an isomorphism (injective + surjective).

2. The map
f : R2 → R
u = (x, y) 7→ f (u) = 2y
is not an isomorphism (f (1, 3) = f (2, 3) but (1, 3) ̸= (2, 3)).

Definition 20. Given f : E → F an isomorphism. Then, the map f −1 : F → E that


associates y in F with the unique x ∈ E such that f (x) = y is called the inverse bijection of
f.

Example 19. The inverse map of


f : R2 → R2
u = (x, y) 7→ f (u) = (x + y, x − y)
is
f −1 : R2 → R2
−a + b a − b
!
v = (a, b) 7→ f −1
(v) = ,
2 2

Proposition 10. Let f : E → F be a bijective map, and f −1 : F → E its inverse bijection.

1. f −1 is bijective, and its inverse bijection is f .

2. f −1 ◦ f = IdE and f ◦ f −1 = IdF .

Definition 21 (Automorphism). A linear map from a vector space E to itself (endomorphism)


that is bijective is called an automorphism. The set of automorphisms of E is denoted
Aut(E).

Example 20. The map


f : R2 → R2
u = (x, y) 7→ f (u) = (x + y, x − y)
is an automorphism.

Proposition 11. Let f be a linear map from E to F with dim E = dim F = n. Then the
following equivalences hold:

f is an isomorphism ⇔ f is surjective ⇔ f is injective ⇔ dim(Im(f )) = dim F

⇔ Im(f ) = F ⇔ dim(ker(f )) = 0 ⇔ ker(f ) = {0E }

20
2.5 Solved Exercises
Exercise 1. Consider the function f defined from R2 to R2 as follows:

f (x, y) = (2x − 4y, x − 2y)

1. Show that f is linear.

2. Determine ker f , Im f , and provide their dimensions.

3. Is f bijective?

Solution

1. For all u = (x, y) and v = (x0 , y0 ) in R2 :

f (u + v) = f ((x, y) + (x0 , y0 )) = f (x + x0 , y + y0 )
= (2(x + x0 ) − 4(y + y0 ), (x + x0 ) − 2(y + y0 ))
= (2x − 4y, x − 2y) + (2x0 − 4y0 , x0 − 2y0 )
= f (u) + f (v)

For all u = (x, y, z) in R2 and λ ∈ K:

f (λu) = f (λ(x, y)) = f (λx, λy)


= (2λx − 4λy, λx − 2λy)
= λ(2x − 4y, x − 2y) = λf (u)

Therefore, f is a linear map.

2. Determine ker f and Im f and provide their dimensions.


The kernel:
ker f = {(x, y) ∈ R2 | 2x − 4y = 0, x − 2y = 0}
ker f = {(x, y) ∈ R2 | x = 2y}
ker f = {(2y, y) | y ∈ R}
ker f = Span{(2, 1)}
And dim ker f = 1.
The image:
Im f = {f (u) | u ∈ R2 }
Im f = {(2x − 4y, x − 2y) | (x, y) ∈ R2 }
Im f = {(2, 1)}
And dim Im f = 1.

dim ker f + dim Im f = dim R2


1+1=2
Thus, the dimensions are consistent.

3. f is not bijective because it is neither injective nor surjective.

21
Exercise 2. Let f : R3 → R3 be defined by f (u) = f (x, y, z) = (−2x + y + z, 4y, x − 2y + z).
1. Show that f is linear.

2. Determine ker f and infer rank f .

3. Calculate Im f .
Solution
1. For all u = (x, y, z) and v = (x0 , y0 , z0 ) in R3 :

f (u + v) = f ((x, y, z) + (x0 , y0 , z0 )) = f (x + x0 , y + y0 , z + z0 )

= (−2(x + x0 ) + (y + y0 ) + (z + z0 ), 4(y + y0 ), (x + x0 ) − 2(y + y0 ) + (z + z0 ))


= (−2x + y + z, 4y, x − 2y + z) + (−2x0 + y0 + z0 , 4y0 , x0 − 2y0 + z0 ) = f (u) + f (v)

For all u = (x, y, z) in R3 and λ ∈ K:

f (λu) = f (λ(x, y, z)) = f (λx, λy, λz)

= (−2λx + λy + λz, 4λy, λx − 2λy + λz)


= λ(−2x + y + z, 4y, x − 2y + z) = λf (u)
Thus, f is a linear map.

2. The kernel:

ker f = {(x, y, z) ∈ R3 | −2x + y + z = 0, 4y = 0, x − 2y + z = 0}

ker f = {(x, y, z) ∈ R3 | −2x + y + z = 0, y = 0, x − 2y + z = 0}


ker f = {(x, 0, 2x) | x ∈ R}
ker f = Span{(1, 0, 2)}
And dim ker f = 1.
The rank:
rank f = dim R3 − dim ker f = 3 − 1 = 2

3. The image:
Im f = {f (u) | u ∈ R3 }
Im f = {(−2x + y + z, 4y, x − 2y + z) | (x, y, z) ∈ R3 }
Im f = {(−2, 4, 1), (1, 0, 1)}
And dim Im f = 2.
Exercise 3. In the vector space R[X] of polynomials, consider the "shift" application D :
R[X] → R[X] such that D(P (X)) = P (X + 1).
1. Is the application D linear?

2. Describe its image (Im D). Is this application surjective?

3. Describe its kernel (ker D). Is this application injective?

4. Is D an isomorphism? If yes, describe its reciprocal linear application.

22
Solution

1. For any pair P, Q of R[X] and any pair α, β of real numbers, we have

D(αP + βQ) = (αP + βQ)(X + 1) = αP (X + 1) + βQ(X + 1)

= αD(P ) + βD(Q)
Thus, D is a linear map.

2. Any polynomial can be written as the image of another under D. Let Q(X) = b0 +
b1 X + . . . + bn X n . Consider the polynomial P (X) = Q(X − 1). Then Q(X) =
P (X + 1) = D(P ). The image of the shift operation is thus the entire space of
polynomials Im D = R[X]. Therefore, D is surjective.

3. We seek polynomials P (X) = a0 + a1 X + . . . + an X n whose image under the shift


operation is zero, i.e., P (X +1) = 0. This means the polynomial Q(X) = P (X +1) = 0
is zero. Thus, P (X) = Q(X − 1) = 0, and ker D = {0}. Therefore, D is injective.

4. The shift application D is both surjective and injective, making it bijective and an
isomorphism. The reciprocal function of D is given by the inverse shift D−1 : R[X] →
R[X] defined as D−1 (Q(X)) = P (X − 1).

Exercise 4. Let E = C 1 (R) be the vector space of continuously differentiable functions, and
φ be a linear map φ : E = C 1 (R) → E = C 1 (R) such that φ(f ) = f ′ .

1. What is the kernel of φ?

2. What is its image?

3. Is φ injective? surjective?

Solution

1. A function f : R → R has its derivative equal to zero if and only if it is constant.


Therefore, the kernel of φ is the set of constant functions.

2. On the other hand, if g is a C 1 (R) function, it has a primitive F , which is also C 1 (R).
We then have φ(F ) = F ′ = g, showing that the image of the differentiation operation
is the space of continuously differentiable functions, i.e., Im φ = E.

3. a) ker φ = {f ∈ C 1 (R) | f ′ = 0}, so φ is not injective.


b) Im φ = E, so φ is surjective.

Exercise 5. Let f be an endomorphism of E, a vector space.

1. Show that ker(f ) ⊆ ker(f 2 ).

2. Show that Im(f 2 ) ⊆ Im(f ).

Correction

1. Let x ∈ ker(f ). This means that f (x) = 0, so f 2 (x) = f (f (x)) = f (0) = 0, which
shows that x ∈ ker(f 2 ). Thus, ker(f ) ⊆ ker(f 2 ).

2. Let y ∈ Im(f 2 ). This means that there exists x ∈ E such that y = f 2 (x) = f (f (x)).
So, y is the image of an element under f , hence y ∈ Im(f ). Thus, Im(f 2 ) ⊆ Im(f ).

23
2.6 Exercices Supplémentaires
Exercise 6. Determine if the following functions are linear:
1. f : R2 → R2
(x, y) 7→ (3x + y, x − y)

2. f : R3 → R3
(x, y, z) 7→ (xy, −x, y)

3. f : R3 → R3
(x, y, z) 7→ (3x + y + z, y − z, x + y)

4. f : R3 [X] → R3
P 7→ (P (−1), P (0), P (1))

Exercise 7. Let f : R3 → R2 be defined for all u = (x, y, z) ∈ R3 by

f : R3 → R2

u 7→ f (u) = (x + y + z, 2x + y − z)

1. Show that f is linear.

2. Determine ker f .

Exercise 8. Let E = R[X], the vector space of polynomials of degree less than or equal to
n, and f : E → E defined by
f (P ) = P + (1 − X)P ′

1. Show that f is an endomorphism of R[X].

2. Determine a basis for Im f .

3. Determine a basis for ker f .

4. Show that Im f and ker f are complementary in R[X].

Exercise 9. Let B = (e1 , e2 , e3 ) be the canonical basis of R3 . Let f : R3 → R3 be an


endomorphism defined by
f (e1 ) = 2e1 + e2 + 3e3
f (e2 ) = e2 − 3e3
f (e3 ) = −2e2 + 2e3

1. For u = (x, y, z) ∈ R3 , determine the image of u under f (calculate f (u)).

2. Let
E = {u ∈ R3 | f (u) = 2u}
F = {u ∈ R3 | f (u) = −u}
Show that E and F are subspaces.

3. Determine a basis for E and a basis for F .

4. Does E ⊕ F = R3 ?

24
Exercise 10. Let f be an endomorphism of a vector space E.

1. Show that Im f ∩ ker f = {0} if and only if ker f = ker f 2 .

2. Show that E = Im f + ker f if and only if Im f = Im f 2 .

3. For endomorphisms f and g on E such that f ◦ g = g ◦ f , show that ker f and Im f


are stable under g.

25

You might also like