Kernel and Range of a linear transformation:

————————————————-
For L : V ! W Kernel of L is
Ker(L) = fu 2 V j L(u) = 0w g
and Ker(L) is a subspace of V:
—————————————
For L : V ! W Range of L is
Range(L) = fw 2 W j L(u) = w for some u 2 V g
and Range(L) is a subspace of W:
—————————————–
If Ker(L) = f0v g then L is one-to-one
If Range(L) = W; then L is onto.
—————————————
Examples: 2 3
u1
1. Let L : R3 ! R2 be given as L(4 u2 5) = u1 u2 u3
u3
Find Ker(L) and a basis for Ker(L): 2 3
u1
Since Ker(L) = fu 2 R3 j L(u) = 0R2 g = f4 u2 5 j u1 u2 u3 =
u3
0 0 g
Then u1 u2 = 0
u3 = 0
Use Gaussian elimination:
1 1 0 0
[Aj0] = in RREF
0 0 1 0
Take u2 = s Then2u1 =3s; u3 = 0
s
Then Ker(L) = f4 s 5 j s 2 Rg
0
2 3
1
Since any vector in Ker(L) can be written as s 4 1 5 then the basis for
0
Ker(L) 82 39
< 1 =
is the set 4 1 5 :
: ;
0
Clearly, dim Ker(L) = 1
Is L one-to-one? 2 3 2 3
s 0
Since Ker(L) = f4 s 5 j s 2 Rg = 6 f4 0 5g
0 0
then L is not one-to-one.
2.Let L : R3 ! R3 be given as L( x y z ) = x + 3z y y+z
(a) Find Ker(L) and a basis for Ker(L):

1
 Ker(L) = f x y z j x + 3z y y+z = 0 0 0 g
Then we have to solve the homogeneous system
x + 3z = 0
y=0
y+z =0
Augmented
2 matrix: 3 2 3
1 0 3 0 1 0 0 0
[Aj0] = 4 0 1 0 0 5, row echelon form: 4 0 1 0 0 5 = [Cj0]
0 1 1 0 0 0 1 0
The solution is x = 0; y = 0; z = 0
The Ker(L) = f 0 0 0 g which is the trivial subspace of R3 :
Basis for Ker(L) is empty set. Then dim Ker(L) = 0:
(b) Find Range(L) and a basis for Range(L)
Range(L) = f x + 3z y y + z j x; y; z 2 Rg:
x + 3z y y+z =x 1 0 0 +y 0 1 1 +z 3 0 1
Then Range(L) = spanf 1 0 0 ; 0 1 1 ; 3 0 1 g
To …nd2 a basis for 3Range(L) write the vectors as colunms of a matrix:
1 0 3
B=4 0 1 0 5
0 1 1
Transform B into a RREF(or REF). Ther colunms containing leading 1s
indicate
the linearly
2 independent
3 vectors. 2 3
1 0 3 1 0 0
B=4 0 1 0 5, row echelon form: 4 0 1 0 5 = I3
0 1 1 0 0 1
All 3 vectors afre linearly independent.
A basis for Range(L) = f 1 0 0 ; 0 1 1 ; 3 0 1
Then dim Range(L) = 3: But dim R3 = 3: Then Range(L) = R3 :
(c) Is L one-to-one? Explain.
Since Ker(L) = f 0 0 0 g then L is one-to-one.
(d) Is L onto? Explain.
Since Range(L) = R3 then L is onto.
Clearly dim Ker(L) + dim Range(L) = 0 + 3 = dim V = dim R3 = 3:
3. 2020-21 Fall
Let L : P2 ! R2 be a linear transformation de…ned as
a b
L(at2 + bt + c) =
a+c
(a) Find Ker(L) and a basis for Ker(L):
0
Ker(L) = fat2 + bt + cjL(at2 + bt + c) = g
0
a b 0
Then = :
a+c 0
We will solve the homogeneous system
a b=0
a+c=0

2
Use Gauss elimination :
1 1 0 0 1 1 0 0
[Aj0] = R1 + R2 ! R2 = [Cj0]
1 0 1 0 ! 0 1 1 0
Cx = 0 : a b = 0
b+c=0
Then a; b are basic and c is free variable.
Take c = s 2 R; then b = s; a = s; where s 2 R:
Ker(L) = f st2 st + sj s 2 Rg:
st2 st + s = s( t2 t + 1)
A basis for Ker(L) is the set f t2 t + 1g:
Then dim Ker(L) = 1:
(b) Find Range(L) and a basis for Range(L):
a b
Range(L) = f j a; b; c 2 Rg:
a+c
a b 1 1 0
=a +b +c
a+c 1 0 1
1 1 0
Range(L) = spanf ; ; g
1 0 1
To …nd a basis for Range(L) :
1 1 0 1 1 0
B= R1 + R2 ! R2 in REF
1 0 1 ! 0 1 1
The columns with leading 1s indicate the linearly
independent vectors which are the …rst two vectors.
1 1
A basis for Range(L) is the set f ; g
1 0
Then dim Range(L) = 2:
Also dim R2 = 2; so Range(L) = R2 :
(c) Is L one-to-one? Explain.
Since Ker(L) = f st2 st + sj s 2 Rg = 6 f0g
then L is not one-to-one.
(d) Is L onto? Explain.

Since Range(L) = R2 then L is onto.