Linear Algebra

Chapter five: Linear Transformations

Mesfin Masre (PhD)

December 24, 2024

1 / 24
Definition of linear transformations and examples

Definition 1
Let V and W be vector spaces over a field F. The map T : V → W is
said be a linear mapping or a linear transformation if it satisfies the
following two properties.
i) T (u + v ) = T (u) + T (v ), for each u, v ∈ V
ii) T (cu) = cT (u), for each u ∈ V and c ∈ F .

Example 2
Show that f : R → R defined by f (x) = x is a linear map.

Solution: i) Let u, v ∈ R. f (u + v ) = u + v = f (u) + f (v ).

ii) ∀u ∈ R and c ∈ F , we have f (cu) = cu = cf (u).
Hence, f is a linear map.

2 / 24
Cont...

Example 3
Let V = Pn (R) over F = R. Then show that f : V → V defined by
f (p(x)) = p ′ (x) is a linear map.

Solution: i) Let p(x), q(x) ∈ V . Then

f (p(x) + q(x)) = (p(x) + q(x))′ = p ′ (x) + q ′ (x) = f (p(x)) + f (q(x)).
ii) ∀p(x) ∈ V and c ∈ F , we have f (cp(x)) = (cp(x))′ = cp ′ (x).
Therefore, f is a linear map.
Example 4
Show that T : R3 → R2 defined by T (x, y , z) = (x + y , y − z) is a linear
transformation.

3 / 24
Properties of linear transformations

Theorem 5
Let V and W be vector spaces over the same field F and T : V → W be
a linear mapping. Then
1 T (0V ) = 0W , where 0V is the zero element of V and 0W is the zero
element of W .
2 T (−u) = −T (u) for each u ∈ V .
3 T (a1 u1 + a2 u2 + · · · + an un ) = a1 T (u1 ) + a2 T (u2 ) + · · · + an T (un )
for each ui ∈ V and ai ∈ F .

Example 6
Let T : R2 → R2 be a linear map satisfying T (1, 1) = (1, 4) and
T (2, −1) = (−2, 3). Then find
i T (3, −1)
ii) T (a, b) for all (a, b) ∈ R2 .
4 / 24
Cont...

Solution: a) To find T (3, −1), we first write (3, −1) as a linear

combination of (1, 1) and (2, −1). Let (3, −1) = α(1, 1) + β(2, −1). Then
(3, −1) = (α + 2β, α − β). This implies that, α + 2β = 3 and
α − β = −1. Solving for α and β, we get α = 31 and β = 43 . That is,
(3, −1) = 31 (1, 1) + 34 (2, −1). By properties of linear map, we have
 
1 4
T (3, −1) = T (1, 1) + (2, −1)
3 3
1 4
= T (1, 1) + T (2, −1)
3 3
1 4
= (1, 4) + (−2, 3)
3 3
−7 16
= ( , ).
3 3
Therefore, T (3, −1) = ( −7 16
3 , 3 ).

5 / 24
Cont...
b) Let (a, b) be an arbitrary element of R2 . To find T (a, b), we first write
(a, b) as a linear combination of (1, 1) and (2, −1). Let
(a, b) = α(1, 1) + β(2, −1). Then (a, b) = (α + 2β, α − β). This implies
that, α + 2β = a and α − β = b. Solving for α and β, we get α = a+2b 3
and β = a−b a+2b a−b
3 . That is, (a, b) = 3 (1, 1) + 3 (2, −1). By properties of
linear map, we have
 
a + 2b a−b
T (a, b) = T (1, 1) + (2, −1)
3 3
a + 2b a−b
= T (1, 1) + T (2, −1)
3 3
a + 2b a−b
= (1, 4) + (−2, 3)
3 3
4b − a 7a + 5b
= ( , ).
3 3
Therefore, T (a, b) = ( 4b−a
3 ,
7a+5b 2
3 ), ∀(a, b) ∈ R .
6 / 24
Kernel and Image of Linear Transformations

Definition 7
Let V and W be vector spaces over F and T : V → W be a linear map.
Then
1 The null space of T or kernel of T , denoted by Ker (T ) is, defined by
Ker (T ) = {u ∈ V |T (u) = 0W }.
2 The range of T or image of T , denoted by Im(T ), is defined as
Im(T ) = {w ∈ W |T (v ) = w for some v ∈ V }.

Theorem 8
Let T : V → W be a linear map. Then
1 Kernel of T is a subspace of V .
2 Image of T is a subspace of T .

7 / 24
Example 9
Find the kernel and image of T : R3 → R3 given by T (x, y , z) = (0, y , z).

Ker (T ) = {(x, y , z) ∈ R3 |T (x, y , z) = (0, 0, 0)}

= {(x, y , z) ∈ R3 |(0, y , z) = (0, 0, 0)}
= {(x, y , z) ∈ R3 |y = 0, z = 0}
= {(x, 0, 0)|x ∈ R}
= {x(1, 0, 0)|x ∈ R}.
That is, Ker (T ) is a subspace of R3 generated by (1, 0, 0).
Im(T ) = {T (x, y , z)|(x, y , z) ∈ R3 }
= {(0, y , z)|y , z ∈ R}
= {y (0, 1, 0) + z(0, 0, 1)|y , z ∈ R}.
That is, Im(T ) is a subspace of R3 generated by vectors (0, 1, 0) and
(0, 0, 1).
8 / 24
Cont...

Example 10
Find the kernel and image of the linear transformation T : R3 → R2
defined by T (x, y , z) = (x + y , y − z).

Solution:

Ker (T ) = {(x, y , z) ∈ R3 |T (x, y , z) = (0, 0)}

= {(x, y , z) ∈ R3 |(x + y , y − z) = (0, 0)}
= {(x, y , z) ∈ R3 x + y = 0, y − z = 0}
= {(x, y , z) ∈ R3 |x = −y = −z}
= {(x, −x, −x)|x ∈ R}
= {x(1, −1, −1)|x ∈ R}

That is, Ker (T ) is a subspace of R3 generated by (1, −1, −1).

9 / 24
Cont...

Im(T ) = {(a, b) ∈ R2 |T (x, y , z) = (a, b) for some (x, y , z) ∈ R3 }

= {(a, b) ∈ R2 |(x + y , y − z) = (a, b), for some (x, y , z) ∈ R3 }
= {(a, b) ∈ R2 |x + y = a, y − z = b}
= {(x + y , y − z)|x, y , z ∈ R}
= {(x + y , 0) + (0, y − z)|x, y , z ∈ R}
= {(x + y )(1, 0) + (y − z)(0, 1)|x, y , z ∈ R}
= {α1 (1, 0) + α2 (0, 1)|α1 = x + y , α2 = y − z}.

Thus, Im(T ) is a subspace of R2 generated by (1, 0) and (0, 1).

10 / 24
Cont...

Theorem 11
Let T : V → W be a linear transformation. Then T is one-to-one if and
only if KerT = {0V }.

Proof: Suppose T is one-to-one. We need to show that KerT contains

only the zero vector. Let u ∈ KerT . Then T (u) = 0W . Since
T (0V ) = 0W , we have T (u) = T (0V ), which implies that u = 0V as T is
one-to-one. Therefore, KerT = {0V }.
Conversely, suppose KerT = {0V }. We need to show that T is
one-to-one. Now, let T (u1 ) = T (u2 ) for some u1 , u2 ∈ V . Then
T (u1 ) − T (u2 ) = T (u1 ) + T (−u2 ) = 0W . Since T is linear, we have
T (u1 − u2 ) = 0W , which implies that u1 − u2 ∈ KerT . Since KerT
contains only the zero vector, we have u1 − u2 = 0V and hence u1 = u2 .
Consequently, T is one-to-one. This completes the proof.

11 / 24
Cont...

Definition 12
Let T : V → W be a linear transformation. Then
a) The dimension of the Kernel (null space) of T is called the nullity of
T.
b) The dimension of the image (range) of T is called the rank of T .

Theorem 13
(Rank-Nullity Theorem) Let V and W be vector spaces over the same
field F and T : V → W be a linear transformation. If V is finite
dimensional vector space, then

dim(V ) = nullity (T ) + rank(T ).

That is, dim(V ) = dim(KerT ) + dim(ImT ).

Proof: Exercise
12 / 24
Cont...

Example 14
Find the nullity and image of the linear transformation T : R3 → R2
defined by T (x, y , z) = (x + y , y − z).

Solution: As we have seen in the above example, the kernel of T is a

subspace of R3 generated by (1, −1, 1). So,
nullity (T ) = dim(Ker (T )) = 1. The image of T is a subspace of R2
generated by (1, 0) and (0, 1). So, rank(T ) = dim(Im(T )) = 2.
Exercise: Find the nullity and rank of the linear transformation
T : R3 → R3 defined by T (x, y , z) = (x, x + 2y , y ).

13 / 24
Cont...

Theorem 15
Let T : V → W be a linear transformation and v1 , v2 , · · · , vn be linearly
independent vectors in V . If Ker (T ) = {0V }, then
T (v1 ), T (v2 ), · · · , T (vn ) are linearly independent vectors in W .

Proof. Let α1 , α2 , · · · , αn be scalars such that

α1 T (v1 ) + α2 T (v2 ) + · · · + αn T (vn ) = 0W . Then by linearity of T , we
have T (α1 v1 + α2 v2 + · · · + αn vn ) = 0W . Hence
α1 v1 + α2 v2 + · · · + αn vn ∈ Ker (T ). But Ker (T ) = {0V }. Thus
α1 v1 + α2 v2 + · · · + αn vn = 0V . Since v1 , v2 , · · · , vn are linearly
independent vectors in V , we have α1 = 0, α2 = 0, · · · , αn = 0. This
completes the proof.

14 / 24
Algebra of linear transformations
The set of linear transformations from a vector space V into itself has
algebraic structure because of ordinary composition of function, provides a
multiplication of such transformations. We shall explore these ideas in this
section.
Notation: We denote the set of all linear transformations from a vector
space V into W over the field F by L(V , W ).

Theorem 16
Let V and W be vector spaces over a field F , and T and S be linear
transformations from V into W and α ∈ F . Then
i) T + S is a linear transformation from V into W .
ii) αT is a linear transformation from V into W .
iii) L(V , W ) is a vector space over F with respect to the operations
▶ (T + S)(v ) = T (v ) + S(v ) and
▶ (αT )(v ) = αT (v ) for all v ∈ V .

Proof: Exercise
15 / 24
Cont...

Theorem 17
Let U, V and W be vector spaces over a field F . Let T : U → V and
S : V → W be linear transformations. Then
i) the composition function S ◦ T , defined by (S ◦ T )(u) = S(T (u)) for
all u ∈ U is a linear transformation.
ii) If T is invertible, then the inverse function T −1 is a linear
transformation.

Example 18
Let T : R3 → R3 be a linear transformation given by
T (x, y , z) = (x, 3y , z). Is T invertible? If so, find a rule for T −1 like the
one which defines T .

16 / 24
Cont...
Solution: We know that KerT = {0} if and only if T is one-to-one.

Ker (T ) = {(x, y , z) ∈ R3 |T (x, y , z) = (0, 0, 0)}

= {(x, y , z) ∈ R3 |(x, 3y , z) = (0, 0, 0)}
= {(x, y , z) ∈ R3 x = 0, 3y = 0, z = 0}
= {(x, y , z) ∈ R3 |x = y = z = 0}
= {(0, 0, 0)}.

This implies that, T is one-to-one. Moreover, from the rank-nullity

theorem, we have dim(R3 ) = dim(KerT ) + dim(ImT ). Since dim(R3 ) = 3
and dim(KerT ) = 0, we have dim(ImT ) = 3. Since ImT is a subspace of
R3 and dim(ImT ) = 3, we have ImT = R3 . Hence T is onto. Since T is
one-to-one and onto, it is bijective map. Therefore, T is invertible.
To find T −1 , let T (x, y , z) = (a, b, c). Then (x, 3y , z) = (a, b, c). This
implies that, x = a, y = b3 , z = c. T (x, y , z) = (a, b, c) if and only if
T −1 (a, b, c) = (x, y , z). Thus, T −1 (a, b, c) = (a, b3 , c).
17 / 24
Cont...

Theorem 19
Let T : V → W be a linear transformation and V be finite dimensional
vector space with dimV = n. Then the following statements are
equivalent.
i) T is one-to-one
ii) If v1 , v2 , · · · , vk are linearly independent vectors in V , then
T (v1 ), T (v2 ), · · · , T (vk ) are linearly independent vectors in ImT .
iii) If {v1 , v2 , · · · , vk } is a basis of V , then {T (v1 ), T (v2 ), · · · , T (vk )} is
a basis of ImT .
iv) dim(ImT ) = n.

Proof: Exercise

18 / 24
Matrix representation of linear transformations

Theorem 20
Let T : Rn → Rm be a linear transformation. Then there exists an m × n
matrix A such that T (X ) = AX for every column vector X in Rn .

Definition 21
In the above theorem, matrix A is called the matrix associated with T , or
A is called the matrix of T .
Question: For a given linear transformation T : V → W , how can we find
the matrix representation of T ?

19 / 24
Cont...
Let T : V → W be a linear transformation and B = {v1 , v2 , . . . , vn } be a basis
for V and B ′ = {w1 , w2 , . . . , wm } be a basis for W . Then to find the matrix
representation of T w.r.t the bases B and B ′ , first find T (vi ) and express it as a
linear combination of wj , for all 1 ≤ i ≤ n and 1 ≤ j ≤ m. Let

T (v1 ) = a11 w1 + a12 w2 + · · · + a1m wm ,

T (v2 ) = a21 w1 + a22 w2 + · · · + a2m wm ,

·
·
·
T (vn ) = an1 w1 + an2 w2 + · · · + anm wm .
Thenthe matrix representationof T w.r.t. the bases B and B ′ is
a11 a21 · · · a2n
 a12 a22 · · · an2 
A= . ..  = (aij )T
n×m .
 
.. ..
 .. . . . 
a1m a2m ··· anm
20 / 24
Cont...
Example 22
Let T : R2 → R2 defined by T (x, y ) = (x, −y ). Find the matrix of T
w.r.t the bases B = {(1, 1), (1, 0)} and B ′ = {(2, 3), (4, 5)}.

Solution:
−9 5
T (1, 1) = (1, −1) = (2, 3) + (4, 5)
2 2
−15 3
T (1, 0) = (1, 0) = (2, 3) + (4, 5).
2 2
Therefore, the matrix of T w.r.t the bases B and B ′ is
 −9 5 T  −9 −15 
A= 2 2 = 2 2 .
−15 3 5 3
2 2 2 2

Example 23
Let T : R3 → R2 defined by T (x, y , z) = (x + 2y + 3z, 4x + 5y + 6z).
Find the matrix of T w.r.t the standard bases
B = {(1, 0, 0), (0, 1, 0), (0, 0, 1)} and B ′ = {(1, 0), (0, 1)}.
21 / 24
Cont...

Solution:
T (1, 0, 0) = (1, 4) = 1(1, 0) + 4(0, 1)
T (0, 1, 0) = (2, 5) = 2(1, 0) + 5(0, 1)
T (0, 0, 1) = (3, 6) = 3(1, 0) + 6(0, 1).
 
1 2 3
Therefore, the matrix of T w.r.t the bases B and B′ is A = .
4 5 6

Exercise 24
Find the matrix representation of the linear transformation T : R3 → R2
defined by T (x, y , z) = (x + y , y + z) w.r.t the bases
B = {(1, 1, 0), (0, 1, 1), (1, 0, 1)} and B ′ = {(1, 2), (3, 1)}.

22 / 24
Eigenvalues and Eigenvectors of a Linear Transformation

Definition 25
Let T : V → V be a linear operator on a vector space V over a field F .
An eigenvalue of T is a scalar λ in F such that there is a non-zero vector
v in V with T (v ) = λv . If λ is an eigenvalue of T , then
a) any vector v in V such that T (v ) = λv is called an eigenvector of T
associated with the eigenvalue λ;
b) the collection of all vectors v of V such that T (v ) = λv is called the
eigenspace associated with λ.

23 / 24
Cont...

Remark: If v and w are eigenvectors associated with eigenvalue λ. Then

i) v + w is also an eigenvector with eigenvalue λ, because:
T (v + w ) = T (v ) + T (w ) = λv + λw = λ(v + w ).
ii) each scalar multiple kv , for k ̸= 0, is also an eigenvector with
eigenvalue λ, because T (kv ) = kT (v ) = k(λv ) = λ(kv ).

Exercise 26
Let T : V → V be a linear operator with KerT ̸= {0}. Prove that every
non zero vector in KerT is an eigenvector of T with eigenvalue 0.

24 / 24