Scribd
Upload
1K views

4 Linear Transformation

The document discusses linear transformations between vector spaces. It begins by defining a linear transformation as a map between vector spaces that satisfies two properties: (1) it is additive and (2) it is homogeneous. Some examples of linear transformations are then provided such as reflection about an axis and projection onto an axis. The document then proves some key properties of linear transformations including that they map the zero vector to the zero vector and preserve scalar multiplication. It also states that a linear transformation is uniquely determined by its action on a basis of the domain vector space. The rank-nullity theorem relating the dimension of the range and nullspace is also discussed.

Uploaded by

alienxx
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
1K views

4 Linear Transformation

The document discusses linear transformations between vector spaces. It begins by defining a linear transformation as a map between vector spaces that satisfies two properties: (1) it is additive and (2) it is homogeneous. Some examples of linear transformations are then provided such as reflection about an axis and projection onto an axis. The document then proves some key properties of linear transformations including that they map the zero vector to the zero vector and preserve scalar multiplication. It also states that a linear transformation is uniquely determined by its action on a basis of the domain vector space. The rank-nullity theorem relating the dimension of the range and nullspace is also discussed.

Uploaded by

alienxx
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 46

Linear Transformations

Rajiv Kumar Math II


Definition : Let U and V be real
vector spaces. A map T: U → V
from U to V is called a linear map,
or Linear transformation, if T
satisfies the following conditions

(i )T ( u + v ) = T ( u ) + T ( v ) for all u , v ∈U
(ii) T(αu) =αT(u) for all u ∈ U
and all real numbers α
Rajiv Kumar Math II
Example
(i) Reflection about x-axis T(x,y)= (x,-y)
(ii) Reflection about y-axis T(x,y)= (-x,y)
(iii)Projection on x-axis T(x,y)=(x,0)
(iv)Projection on y-axis T(x,y)=(0,y)

Rajiv Kumar Math II


Example: Let a map T: V2→V3 be defined by

T(x1, x2) = (x1, x2 ,x1 x2). Check whether T is

linear.

Rajiv Kumar Math II


T (α(x1,x2)) = T(αx1, αx2)
= (αx1, αx2 , αx1 αx2 )
= (αx1, αx2 , α2x1x2)
= α(x1, x2 αx1x2 )
≠α(x1, x2 ,x1 x2).
Hence T is not linear.
Rajiv Kumar Math II
Example: Let C[-1, 1] be the vector space of
real valued continuous functions defined on
interval [-1, 1],
T(f(x)) = x f ′(x) for all f(x) ∈C[-1, 1].
show that T is a linear map.

Rajiv Kumar Math II


Let f(x), g(x) ∈C[-1,1]
T( f (x) + g(x))
= x (f(x)+g(x)' = x(f '(x)+g'(x) )
= x f'(x) +x g'(x) =T( f (x) )+ T(g(x))

T (αf(x)) = x [αf (x)]' =x(αf '(x))= α T (f(x))

Rajiv Kumar Math II


Theorem: Let T: U→V be a linear map. then
• T(0U)=0V,
• T(-u) = -T(u), and
• T(α1u1+…+αnun)=
α1T(u1)+…+ α
nT(un).
In other words, a linear map T transforms the
0U into the 0V
and the negative of every u∈U into the
negative of T(u) in V.
Rajiv Kumar Math II
Proof of (i) As T is linear
T (αu ) = αT ( u )

If α = 0, ⇒ T ( 0u ) = 0T ( u ) ⇒ T ( 0u ) = 0v

Rajiv Kumar Math II


induction on ‘n’, starting from the fact that

T(αu) = αT(u) and using the property

T(αu1+βu2)=T(αu1)+T(βu2)=αT(u1)+ βT(u2).

Rajiv Kumar Math II


In view of (iii), we get a technique of defining a
linear transformation
T on a finite-dimensional vector space U.
Suppose B ={u1,u2,…..,un}is a basis for U. Then
any vector u∈U can be expressed uniquely in
the form
u = α1u1+ α2u2+…+ αnun.
Rajiv Kumar Math II

So, if T:U→V is a linear map, then


T(u) =T(α1u1+α2u2+…+ αnun)

= α1T(u1)+ α2T(u2)+…+ αnT(un).

Thus, T(u) is known as soon as T(u1),

T(u2),…T(un) are known,

Rajiv Kumar Math II


Theorem: A linear transformation T is
completely determined by its values on the
elements of a basis. Precisely, if B={u1,u2,…

un} is a basis for U and

v1, v2, ….vn be n vectors


(not necessarily distinct) in V
Rajiv Kumar Math II
then there exists a unique linear
transformation T:U→V ,Such that

T(ui)=vi for 1 ≤i ≤n

Rajiv Kumar Math II


Q.2 Determine whether there exists a
linear map satisfying conditions.
If it does exist give general formula.
• T : V2→V2 T(1,1)=(3,0) ,T(-1,1)= (0,1)
(ii) T : V2→V2
T(0,1)= (3,4) , T(3,1)=(2,2) & T(3,2)=(5,7)
(iii) T : P3→P3 , such that T(1+x)=1+x
T(2+x)= x+3x3 & T(x2)=0
(iv) T : V3→V3
T(0,1,2)= (3,1,2) , T(1,1,1)=(2,2,2)
Rajiv Kumar Math II
What is general idea in doing such a problem?
(ii) If set of elements on which
transformation T is defined is independent
then transformation T exist.

(ii) If set of elements on which


transformation T is defined is dependent
then transformation T may or may not
exist.

Rajiv Kumar Math II


T : V2→V2 , T(1,1)=(3,0) ,T(-1,1)= (0,1)
{(1,1), (-1,1)} is linearly independent. , hence
such a linear transformation exist .

{(1,1), (-1,1)} is basis of V2

To find linear transformation we


express (x,y) as linear combination of
{(1,1), (-1,1)}.
Rajiv Kumar Math II
(x,y) = α(1,1)+β(-1,1)= (α-β,α+β)
⇒ α-β=x & α+β=y
Solving we get
α=(x+y)/2 & β = (y-x)/2
(x,y) = ((x+y)/2) (1,1)+ ((y-x)/2) (-1,1)
T(x,y) = ((x+y)/2) T(1,1)+ ((y-x)/2) T(-1,1),

T(x,y) = ((x+y)/2) (3,0)+ ((y-x)/2) (0,1)


= ((3x+3y)/2, (y-x)/2)
Rajiv Kumar Math II
Definition: Let T:U→V be a
linear map then Range of T
R(T)= {v∈ V | v=Tu, for some u
∈U}
The kernel(null space) of T is the set
ker(T) =N(T)={u∈U|T(u)=0v}.
n other words, N(T) is the set of all
those elements in U that are mapped
Rajiv Kumar Math II

by T into the zero vector of V.


Theorem: Let T:U→V be a linear
map.
. R(T) is a subspace of V
i. N(T) is a subspace of U
ii.T is one-one iff N(T) is the zero
subspace of U.
v If [{u1,u2,….,un}]=U,
then R(T) =[{T(u1),T(u2),…T(un)}].
V. If U is finite-dimensional, then
dim (R(T) )≤ dim U.
Rajiv Kumar Math II
Proof: (i) Let v1, v2∈R(T). Then there

exist vectors u1, u2 in U such that

T(u1)=v1 and T(u2)=v2. So


1+v2=T(u1)+T(u2)=T(u1+u2)

as T is linear) But u1+u2∈U, since U is

vector space. Hence, v1+v2 is the image

n element of U. So v +v ∈R(T).
Rajiv Kumar Math II
In the same way,αT(u1)=T(αu1),

since T is linear. But αu1∈U, as U

is a vector space. Hence, αv1∈

R(T).Thus R(T) is a subspace of


V. Rajiv Kumar Math II
(ii) To prove N(T) is a subspace
of U

Let u, v ∈N(T). Then T(u)= 0v

and T(v)=0v, Now

T(u + v)= T(u)+ T(v)=0v+0v =0v

⇒ u + v∈N(T) Rajiv Kumar Math II


(iii) T is one-one iff N(T) is the
zero subspace of U.

Proof :
u ∈N(T) we want to prove u
=0U .
T(u)= 0v =T(0U) , as T is linear
⇒ u= 0U as T is one-one .
⇒ N(T) is zero subspace of
U Rajiv Kumar Math II
If N(T) =0U then T is one one
T(u)= T(v) ⇒ T(u-v)= 0V
⇒u-v∈ N(T) ⇒ u-v= 0U
(as N(T )=0U )
i.e. u=v ⇒ T is one-one.

Rajiv Kumar Math II


(iv) If [{u1,u2,….,un}]=U,
then R(T)
=[{T(u1),T(u2),…T(un)}].
We need to prove
[{T(u1),T(u2),…T(un)}]⊆ R(T)
………..(I)
&

R(T)⊆[{T(u1),T(u2),…T(un)}]……
(II) Rajiv Kumar Math II
T(ui)∈R(T) ,∀i, 1≤i≤n

⇒[{T(u1),T(u2),…T(un)}]⊆ [R(T)]
=R(T)
As R(T) is the subspace

Rajiv Kumar Math II


v∈R(T) ⇒ ∃u∈U suct that v =Tu

u∈U⇒ u =α1u1 +,... + α nun


As [{u1,u2,….,un}]=U,

T (u ) = α1T ( u1 ) + α 2T ( u2 ) +,...,+α nT ( un )

⇒R(T)⊆[{T(u1),T(u2),…T(un)}]
……(II)
Rajiv Kumar Math II
v.If U is finite-dimensional, then
dim (R(T) )≤ dim U.
Let {u1,u2,…un } be a basis of
U
Then [{u1,u2,…un }]= U
⇒R(T) =[{T(u1),T(u2),…T(un)}]
dim (R(T))
=dim [{T(u1),T(u2),…T(un)}]
≤ n =dim U.
Rajiv Kumar Math II
Example Let T: V4 → P3 be a linear
map from defined as
T ( x1, x2 , x3 , x4 ) = x1 + ( x2 − x3 ) x + ( x1 − x3 ) x 3

(i) Find N(T) and a basis of N(T)

(ii) Find R(T) and a basis of R(T)

Rajiv Kumar Math II


{
N (T ) = ( x1, x2 , x3 , x4 ) x1 + ( x2 − x3 ) x + ( x1 − x3 ) x 3 = 0 P
3
}

( x1, x2 , x3 , x4 ) | x1 = 0, x2 − x3 = 0 
= 
 & x1 − x3 = 0 
= { (0,0,0, x4) | x4 is any real number }
= [(0,0,0,1)]
Basis of N(T) is {(0,0,0,1)}
Rajiv Kumar Math II
R (T ) = [{T (1,0,0,0), T (0,1,0,0), T (0,0,1,0), T (0,0,0,1)}]
As {(1,0,0,0), (0,1,0,0), (0,0,1,0), (0,0,0,1)}
is a basis of V4.

T(1,0,0,0) = 1+x3

T(0,1,0,0) = 0+ (1-0)x =x
T(0,0,1,0) = 0+ (0-1)x +(0-1)x3

T(0,0,0,1) = 0V Rajiv Kumar Math II


Hence R(T) = [{1+x3 ,x, -x-x3, 0V
}]
Hence R(T) = [{1+x3,x, -x-
x3}]
Set {1+x3 ,x, -x-x3} is independe
In P3, (prove that )

Hence {1+x3 ,x, -x-x3} is a basis


of R(T)
Rajiv Kumar Math II
Rank of a linear transformation T
T : U→V, U finite dimensional
Definition : dim R(T) is
called rank(T) and is
written as r(T).

Rajiv Kumar Math II


Nullity of a linear transformation T
T : U→V, U finite dimensional
Definition : dim N(T) is
called nullity(T) and is
written as n(T).

Rajiv Kumar Math II


Rank Nullity Theorem

T : U → V , a linear
transformation & U finite
dimensional
Then r(T) +n(T)= dim(U)

Rajiv Kumar Math II


Example : Let T: V3 → V4 be a linear map,
defined by
T ( x1, x2 , x3 ) =
( x1, x2 − x3 , x1 − x2 + x3 , x1 + x2 − x3 ) .

Verify rank nullity theorem. .

Rajiv Kumar Math II


N (T ) = {( x1, x2 , x3 ) : T ( x1, x2 , x3 ) = (0,0,0,0)}
⇒ ( x1, x2 − x3 , x1 − x2 + x3 , x1 + x2 − x 3 )
= (0,0,0,0)
⇒ x1 = 0,
x2 − x3 = 0,
x1 − x2 + x3 = 0
x1 + x2 − x 3 = 0,
Rajiv Kumar Math II
Solving system of four
equations in three
unknowns we get
x1 =0 & x3 = x2 .

Rajiv Kumar Math II


N(T) ={(0,x3, x3) : x3 is arbitrary}
=[{(0,1,1)}]

R(T)=
{ (x1,x2 - x3,x1 -x2 +x3, x1 +x2 -x3):(x1,x2, x3)∈
V3}
= [{(1,0,1,1), (0,1,-1,1),(0,-1,1,-1)}]
What is dim (R(T)) ?

Rajiv Kumar Math II


R(T)= [{(1,0,1,1), (0,1,-1,1),(0,-1,1,-
1)}]
To find dimension of R(T) we
find
Linearly independent subset A
of set
S= {(1,0,1,1), (0,1,-1,1),(0,-1,1,-
1)}
Such that [A]=[S]=R(T)
Rajiv Kumar Math II
Theorem : If U & V are finite
dimensional vector spaces of the
same dimension , then a linear map
T:U→ V is one-one iff it is onto.

Rajiv Kumar Math II


Question : If T : V4→ V3
is the linear transformation
defined as
T(1,0,0,0)= (-1,0,1) ,
T(0,1,0,0)= (3,-1,0)
T(0,0,1,0)= (2,-1,1)
T(0,0,0,1)= (1,0,-1)
find ker(T) & r(T) ?
Rajiv Kumar Math II
To find ker (T) we must find
transformation T defined for
any element of V4.
T(x1,x2, x3 ,x4)=
(-x1+3x2 +2x3+ x4, -x2 -x3, x1 +x3
-x4 )

Rajiv Kumar Math II


T(x1,x2, x3 ,x4)=(-x1+3x2 +2x3+ x4, -x2 -x3, x1 +x3 -x4
)

ker(T ) = {( x1, x2 , x3 , x4 ) : T ( x1, x2 , x3 , x4 ) = (0,0,0)}

-x1+3x2 +2x3+ x4 =0
-x2 -x3, =0
x1 +x3 -x4 =0
Solving system of three
equations in four unknowns
we get x1 = -x3 +x4 , x2 =-x3`
Rajiv Kumar Math II
Question : If T : V4→ V3
is the linear transformation
defined as
T(1,0,0,0)= (-1,0,1) ,
T(0,1,0,0)= (3,-1,0)
T(0,0,1,0)= (2,-1,1)
T(0,0,0,1)= (1,0,-1)
find R(T) & n(T) ?
Rajiv Kumar Math II

You might also like