Linear Algebra and Its Applications

Sixth Edition, Global Edition

Session 5
Linear
Transformations

Slide - 1
Section 1.8: Introduction To Linear
Transformations

Slide - 2
Linear Transformations
• A transformation (or function or mapping) T from ℝ𝑛 to ℝ𝑚
𝑛
is a rule that assigns to each vector x in ℝ a vector T (x) in ℝ𝑚 .

• The set ℝ𝑛 is called domain of T, and ℝ𝑚 is called the codomain of T.

• The notation 𝑇: ℝ𝑛 → ℝ𝑚 indicates that the domain of T is ℝ𝑛

and the codomain is ℝ𝑚 .
𝑛 𝑚
• For x in ℝ , the vector T (x) in ℝ is called the image of x (under the
action of T ).

• The set of all images T(x) is called the range of T. See the figure on the
next slide.

Slide - 3
Matrix Transformations (1 of 7)

𝑛
• For each x in ℝ , T (x) is computed as Ax, where A is an m  n matrix.

• For simplicity, we denote such a matrix transformation by x Ax.

𝑛
• The domain of T is ℝ when A has n columns and the codomain of T is
ℝ𝑚 when each column of A has m entries.

Slide - 4
Matrix Transformations (2 of 7)
• The range of T is the set of all linear combinations of the columns of A,
because each image T (x) is of the form Ax.

• Example 1: Let

 1 − 3  3  3
  2
A= 3 
5 , u =   , b =  2 , c = 2 ,
   −1    
 −1 7   −5  5

and define a transformation 𝑇: ℝ2 → ℝ3 by T (x) = Ax,

so that
 1 − 3  x1 − 3 x2 
 
T (x) = Ax =  3 5  1  =  3 x1 + 5 x2  .
x
   x2   
 −1 7   − x1 + 7 x2 

Slide - 5
Matrix Transformations (3 of 7)
a. Find T(u), the image of u under the transformation T.

b. Find an x in ℝ2 whose image under T is b.

c. Is there more than one x whose image under T is b?

d. Determine if c is in the range of the transformation T.

Slide - 6
Matrix Transformations (4 of 7)
• Solution:

a. Compute
 1 −3  5
   2  
T (u) = Au = 3 5   = 1 .
   −1  
 −1 7   −9 

b. Solve T (x) = b for x. That is, solve Ax = b, or

 1 −3  3
 3 5  x1  =  2  . ----(1)
   x2   
 −1 7   −5

Slide - 7
Matrix Transformations (5 of 7)
• Row reduce the augmented matrix:

 1 −3 3  1 −3 3  1 −3 3  1 0 1.5
 3 5 2 0 14 −7  0 1 −.5 0 1 −.5
       
 −1 7 −5 0 4 −2  0 0 0  0 0 0 
----(2)

 1.5
• Hence x1 = 1.5, x2 = −.5, and x =   .
 −.5
• The image of this x under T is the given vector b.

Slide - 8
Matrix Transformations (6 of 7)
c. Any x whose image under T is b must satisfy equation (1).
– From (2), it is clear that equation (1) has a unique solution.
– So there is exactly one x whose image is b.
d. The vector c is in the range of T if c is the image of some x in ℝ2 ,
that is, if c = T (x) for some x.
– This is another way of asking if the system Ax = c
is consistent.

Slide - 9
Matrix Transformations (7 of 7)
– To find the answer, row reduce the augmented matrix.

 1 −3 3  1 −3 3  1 −3 3  1 −3 3
 3 5 2 0 14 −7  0 1 2 0 1 2
       
 −1 7 5 0 4 8 0 14 −7  0 0 −35

– The third equation, 0 = −35, shows that the system is inconsistent.

– So c is not in the range of T.

Slide - 10
Shear Transformation (1 of 2)
 1 3
• Example 2: Let A =   . The transformation
0 1
𝑇: ℝ2 → ℝ2 defined by T ( x) = Ax is called a shear
transformation.

• It can be shown that if T acts on each point in the 2  2

square shown in the figure on the next slide, then the set
of images forms the shaded parallelogram.

Slide - 11
Shear Transformation (2 of 2)

• The key idea is to show that T maps line segments onto

line segments and then to check that the corners of the
square map onto the vertices of the parallelogram.
0
• For instance, the image of the point u =   is
2
 1 3  0   6 
T (u) =     =  ,
0 1  2   2 
Slide - 12
Linear Transformations (1 of 5)
2  1 3  2   8 
and the image of   is     =  .
2 0 1  2   2 

• T deforms the square as if the top of the square were pushed to the
right while the base is held fixed.

• Definition: A transformation (or mapping) T is linear if:

i. T (u + v ) = T (u) + T ( v) for all u, v in the domain of T;

ii. T (cu) = cT (u) for all scalars c and all u in the domain of T.

Slide - 13
Linear Transformations (2 of 5)
• Linear transformations preserve the operations of vector addition
and scalar multiplication.
• Property (i) says that the result T (u + v ) of first adding u and v in
ℝ𝑛 and then applying T is the same as first applying T to u and v and
then adding T (u) and T (v) in ℝ𝑚 .

• These two properties lead to the following useful facts.

• If T is a linear transformation, then

T ( 0) = 0 ----(3)

Slide - 14
Linear Transformations (3 of 5)
and T (cu + dv ) = cT (u) + dT ( v) ----(4)
for all vectors u, v in the domain of T and all scalars c, d.
• Property (3) follows from condition (ii) in the definition, because
T (0) = T (0u) = 0T (u) = 0

• Property (4) requires both (i) and (ii):

T (cu + dv ) = T (cu) + T (dv) = cT (u) + dT ( v)

• If a transformation satisfies (4) for all u, v and c, d, it must be linear.

• (Set c = d = 1 for preservation of addition, and set d = 0 for

preservation of scalar multiplication.)

Slide - 15
Linear Transformations (4 of 5)
• Repeated application of (4) produces a useful generalization:
T (c1 v1 + ... + c p v p ) = c1T ( v1 ) + ... + c pT ( v p ) ----(5)

• In engineering and physics, (5) is referred to as a

superposition principle.

• Think of v1 , , v p as signals that go into a system and T

T ( v1 ) , , T ( v p ) as the responses of that system to the signals.

Slide - 16
Linear Transformations (5 of 5)
• The system satisfies the superposition principle if
whenever an input is expressed as a linear combination of
such signals, the system’s response is the same linear
combination of the responses to the individual signals.
• Given a scalar r, define 𝑇: ℝ2 → ℝ2 by T (x) = rx.
• T is called a contraction when 0  r  1
and a dilation when r  1.

Slide - 17
Section 1.9: The Matrix of A Linear
Transformation

Slide - 18
The Matrix of A Linear Transformation (1 of 4)
• Theorem 10: Let 𝑇: ℝ𝑛 → ℝ𝑚 be a linear transformation.
Then there exists a unique matrix A such that

T ( x ) = Ax for all x in ℝ𝑛n

• In fact, A is the m  n matrix whose jth column is the vector

T ( e j ) , where e j is the jth column of the identity matrix in ℝ𝑛

A = T ( e1 )  T ( e1 )  ( 3)

Slide - 19
The Matrix of A Linear Transformation (2 of 4)
• Proof: Write

x = I n x = e1 ... e n  x = x1e1 + ...+ x n e n

and use the linearity of T to compute

T ( x ) = T ( x1e1 +  + xn e n ) = x1T ( e1 ) +  + xnT ( e n )

 x1 
 
 
= T ( e1 )  T ( e1 )    = Ax
 
 
 xn 

Slide - 20
The Matrix of A Linear Transformation (3 of 4)
• The matrix A in (3) is called the standard matrix for the
linear transformation T.
• We know now that every linear transformation from
ℝ𝑛n to ℝ𝑚m can be viewed as a matrix transformation,
and vice versa. The term linear transformation focuses on
a property of a mapping, while matrix transformation
describes how such a mapping is implemented, as the
example on the next slide illustrates.

Slide - 21
The Matrix of A Linear Transformation (4 of 4)
• Example 2: Find the standard matrix A for the
dilation transformation T ( x ) = 3x, for x in ℝ22 .

• Solution: Write

Slide - 22
Geometric Linear Transformations of
Double Lined R Squared (1 of 6)
• Tables 1-4 illustrate other common geometric linear
transformations of the plane.
Table 1 Reflections

Transformation Image of the Unit Square Standard Matrix

1 0 
a 2 by 2 matrix with elements as follows.
Row 1. 1, 0. Row 2. 0, negative 1.

Reflection through the

x1 -axis 0 −1
x sub 1
 
A graph plots an image of a gold fish in a x sub 1 x sub 2 plane with its bottom left corner at the origin and the bottom right corner at a point labeled, a 2 by 1 matrix with elements as follows.
Row 1. 1. Row 2. 0. The image is inverted , with the bottom left corner of the inverted image at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 0. Row 2. negative 1.

Slide - 23
Geometric Linear Transformations of
Double Lined R Squared (2 of 6)
• Table 1 continued:

Transformation Image of the Unit Square Standard Matrix

 −1 0 
a 2 by 2 matrix with elements as follows. Row 1. 0, 1. Row 2. 1, 0.

Reflection through
the x2 –axis  0 1
x sub 2
 

A graph plots an image of a gold fish in a x sub 1 x sub 2 plane with its bottom left corner at the origin and the top left corner at a point labeled, a 2 by 1 matrix with elements as follows.
Row 1. 0. Row 2. 1. The image is mirror imaged, with the bottom left corner of the mirrored image at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. negative 1. Row 2, 0.

0 1 
a 2 by 2 matrix with elements as follows. Row 1. 0, 1. Row 2. 1, 0.

Reflection through
the line x2 = x1 1 0
x sub 2 = x sub 1
 
A graph plots an image of a gold fish in a x sub 1 x sub 2 plane with its bottom left corner at the origin and the bottom right corner at a point labeled, a 2 by 1 matrix with elements as follows.
Row 1. 1. Row 2. 0. The image is diagonally inverted , with the top left corner of the inverted image at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 0. Row 2. 1. A text reads, x sub 2 = x sub 1.

Slide - 24
Geometric Linear Transformations of
Double Lined R Squared (3 of 6)
• Table 1 continued:
Transformation Image of the Unit Square Standard Matrix
 0 −1
a 2 by 2 matrix with elements as follows. Row 1. 0, negative 1. Row 2. negative 1, 0.

Reflection through
the line x2 = − x1  −1 0 
 
x sub 2 = negative x sub 1

A graph plots an image of a gold fish in a x sub 1 x sub 2 plane with its bottom left corner at the origin and the top left corner at a point labeled, a 2 by 1 matrix with elements as follows.
Row 1. 0. Row 2. 1. The image is diagonally inverted , with the top left corner of the inverted image at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. negative 1. Row 2, 0, and the bottom right corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1., 0. Row 2, negative 1. A text reads, x sub 2 = negative x sub 1.

 −1 0 
a 2 by 2 matrix with elements as follows. Row 1. negative 1, 0. Row 2. 0, negative 1.

Reflection through
the origin  0 −1
 

A graph plots an image of a gold fish in a x sub 1 x sub 2 plane with its bottom left corner at the origin and the top left corner at a point labeled, a 2 by 1 matrix with elements as follows.
Row 1. 0. Row 2. 1. The image is diagonally inverted but its diagonal sides are mirrored, with the top left corner of the inverted image at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. negative 1. Row 2, 0, and the bottom right corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1., 0. Row 2, negative 1.

Slide - 25
Geometric Linear Transformations of
Double Lined R Squared (4 of 6)
Table 2 Contractions and Expansions

Transformation Image of the Unit Square Standard Matrix

a 2 by 2 matrix with elements as follows. Row 1. k, 0. Row 2. 0, 1.

Horizontal  k 0
contraction and 0 1 
 
expansion
Two graphs. First graph plots an image of a gold fish in a x sub 1 x sub 2 plane with its bottom left corner at the origin, the top left corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 0. Row 2. 1, and its bottom right corner at a point labeled, a 2 by 1 matrix with elements as follows.
Row 1. k. Row 2. 0. The image is shrunk vertically. A text reads, k is greater than 0 and is less than 1. Second graph plots an image of a gold fish in a x sub 1 x sub 2 plane with its bottom left corner at the origin, the top left corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 0. Row 2. 1, and its bottom right corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. k. Row 2. 0. The image is expanded vertically. A text reads, k is greater than 1.

a 2 by 2 matrix with elements as follows. Row 1. 1, 0. Row 2. 0, k.

Vertical 1 0 
contraction and 0 k 
 
expansion
Two graphs. First graph plots an image of a gold fish in a x sub 1 x sub 2 plane with its bottom left corner at the origin, t he top left corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 0. Row 2. k, and its bottom right corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 1. Row 2. 0. The image is shrunk horizontally.
Second graph plots an image of a gold fish in a x sub 1 x sub 2 plane with its bottom left corner at the origin, the top left corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 0. Row 2. k, and its bottom right corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 1. Row 2. 0. The image is expanded horizontally.

Slide - 26
Geometric Linear Transformations of
Double Lined R Squared (5 of 6)
Table 3 Shears

Transformation Image of the Unit Square Standard Matrix

a 2 by 2 matrix with elements as follows. Row 1. 1, k. Row 2. 0, 1.

Horizontal shear 1 k 
0 1 
 
Two graphs. First graph plots an image of a bird in a x sub 1 x sub 2 plane with its bottom left corner at the origin, the top left corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. k. Row 2. 1, and its bottom right corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. k. Row 2. 0. The image is skewed leftward.
A text reads, k is less than 0. Second graph plots an image of a bird in a x sub 1 x sub 2 plane with its bottom left corner at the origin, the top left corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. k. Row 2. 1, and its bottom right corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. k. Row 2. 0. The image is skewed rightward. A text reads, k is greater than 0. Dashed lines extend from the point, a 2 by 1 matrix with elements as follows. Row 1. k. Row 2. 1, to the point k on the x axis.

a 2 by 2 matrix with elements as follows. Row 1. 1, 0. Row 2. k, 1.

Vertical shear 1 0
k 1 
 
Two graphs. First graph plots an image of a bird in a x sub 1 x sub 2 plane with its bottom left corner at the origin, the top left corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 0. Row 2. 1, and its bottom right corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 1. Row 2. k. The image is skewed downward.
A text reads, k is less than 0. Second graph plots an image of a bird in a x sub 1 x sub 2 plane with its bottom left corner at the origin, the top left corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 0. Row 2. 1, and its bottom right corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 1. Row 2. k. The image is skewed uptward. A text reads, k is greater than 0. Dashed lines extend from the point, a 2 by 1 matrix with elements as follows. Row 1. 1. Row 2. k, to the point k on the y axis.

Slide - 27
Geometric Linear Transformations of
Double Lined R Squared (6 of 6)
Table 4 Projections
Transformation Image of the Unit Square Standard Matrix

1 0
a 2 by 2 matrix with elements as follows. Row 1. 1, 0. Row 2. 0, 0.

Projection onto
the x1 -axis 0 0
x sub 1
 
A graph plots an image of a gold fish in a x sub 1 x sub 2 plane with its bottom left corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 0. Row 2. 0, and its bottom right corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 1. Row 2. 0. The image is shrunk horizontally along the x axis.

0 0
a 2 by 2 matrix with elements as follows. Row 1. 0, 0. Row 2. 0, 1.

Projection onto
the x2 -axis 0 1 
x sub 2
 
A graph plots an image of a gold fish in a x sub 1 x sub 2 plane with its bottom left corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 0. Row 2. 0, and its top left corner at a point labeled, a 2 by 1 matrix with elements as follows. Row 1. 0. Row 2. 1. The image is shrunk vertically along the y axis.

Slide - 28
Existence And Uniqueness Questions (1 of 8)
• Definition: A mapping T : ℝ𝑛n → ℝm𝑚 is said to be onto
ℝ𝑛 if each b in ℝ𝑚 is the image of at least one x in ℝ𝑛 .

Figure 3 Is the range of T all of ℝ𝑚 ?

Slide - 29
Existence And Uniqueness Questions (2 of 8)
• Equivalently, T is onto ℝ𝑚 when the range of T is all of
the codomain ℝ𝑚 . That is, T maps ℝ𝑛n onto ℝ𝑚m if, for each
b in the codomain ℝ𝑚 , there exists at least one solution of
T (x) = b. “Does T map ℝ𝑛n onto ℝ𝑚m ?” is an existence
question. The mapping T is not onto when there
Is some b in ℝ𝑚 for which the equation T (x) = b
has no solution.

Slide - 30
Existence And Uniqueness Questions (3 of 8)
• Definition: A mapping T : ℝ𝑛n → ℝ𝑚m is said to be one-to
-one if each b in ℝ𝑚 is the image of at most one x
in ℝ𝑛 .

Figure 4 Is every b the image of at most one vector?

Slide - 31
Existence And Uniqueness Questions (4 of 8)
• Example 4: Let T be the linear transformation whose
standard matrix is

1 −4 8 1 
A = 0 2 −1 3
 
0 0 0 5

• Does T map ℝ44 onto ℝ3 ? Is T a one-to-one mapping?

Slide - 32
Existence And Uniqueness Questions (5 of 8)
• Solution: Since A happens to be in echelon form, we can
see at once that A has a pivot position in each row.
By Theorem 4 in Section 1.4, for each b in ℝ33 , the
equation Ax = b is consistent. In other words, the linear
transformation T maps ℝ 44
( its domain ) onto ℝ33
.

• However, since the equation Ax = b has a free variable

(because there are four variables and only three basic
variables), each b is the image of more than one x. This
is, T is not one-to-one.

Slide - 33
Existence And Uniqueness Questions (6 of 8)
• Theorem 11: Let T : ℝ𝑛n → ℝ𝑚
m
be a linear transformation. Then T is one-
to-one if and only if the equation T (x) = 0 has only the trivial solution.

• Proof: Since T is linear, T (0) = 0. If T is one-to-one, then the equation

T (x) = 0 has at most one solution and hence only the trivial solution.

• If T is not one-to-one, then there is a b that is the image of at least

two different vectors in ℝ𝑛 --say, u and v. That is
T (u) = b and T ( v ) = b . But then, since T is linear,

T ( u − v ) = T ( u ) − T ( v ) = b − b =0

Slide - 34
Existence And Uniqueness Questions (7 of 8)
• The vector u − v is not zero, since u  v. Hence the equation T (x) = 0
has more than one solution. So, either the two conditions in the
theorem are both true or they are both false.

• Theorem 12: Let T: ℝ𝑛→ ℝ𝑚 be a linear transformation and let A

n m

be the standard matrix for T. Then:

a) T maps ℝ𝑛n onto ℝ𝑚

m
if and only if the columns of A span ℝ𝑚 ,
b) T is one-to-one if and only if the columns of A are linearly independent.

Slide - 35
Existence And Uniqueness Questions (8 of 8)
• Proof:

a) By Theorem 4 in Section 1.4, the columns of A span ℝ𝑚

if and only if for each b in ℝ𝑛 the equation Ax = b is consistent—
in other words, if and only if for every b, the equation T ( x) = b
has at least one solution. This is true if and only if T maps ℝ𝑛n ontoℝ𝑚m ..

b) The equations T (x) = 0 and Ax = 0 are the same except for notation.
So, by Theorem 11, T is one-to-one if and only if Ax = 0 has only the
trivial solution. This happens if and only if the columns of A are
linearly independent.

Slide - 36