For the time being, well look at ranks and nullity

of transformations. Well come back to these topics

again when we interpret our results for matrices.
The above theorem implies this corollary.

Kernel, image, nullity, and rank

Math 130 Linear Algebra

Corollary 4. Let V W and W X. Then

D Joyce, Fall 2013

nullity(T ) nullity(U T )

Definition 1. Let T : V W be a linear trans- and

formation between vector spaces. The kernel of T ,
rank(U T ) rank(U ).
also called the null space of T , is the inverse image
of the zero vector, 0, of W ,
Systems of linear equations and linear trans1
formations. Weve seen how a system of m linker(T ) = T (0) = {v V | T v = 0}.
ear equations in n unknowns can be interpreted as a
Its sometimes denoted N (T ) for null space of T .
single matrix equation Ax = b, where x is the n1
The image of T , also called the range of T , is the column vector whose entries are the n unknowns,
set of values of T ,
and b is the m 1 column vector of constants on
the right sides of the m equations.
T (V ) = {T (v) W | v V }.
We can also interpret a system of linear equations
This image is also denoted im(T ), or R(T ) for range in terms of a linear transformation. Let the linear
of T .
transformation T : Rn Rm correspond to the
Both of these are vector spaces. ker(T ) is a sub- matrix A, that is, T (x) = Ax. Then the matrix
space of V , and T (V ) is a subspace of W . (Why? equation Ax = b becomes
Prove it.)
We can prove something about kernels and images directly from their definition.
T

T (x) = b.
Solving the equation means looking for a vector
x in the inverse image T 1 (b). It will exist if and
only if b is in the image T (V ).
When the system of linear equations is homogeneous, then b = 0. Then the solution set is the
subspace of V weve called the kernel of T . Thus,
kernels are solutions to homogeneous linear equations.
When the system is not homogeneous, then the
solution set is not a subspace of V since it doesnt
contain 0. In fact, it will be empty when b is not in
the image of T . If it is in the image, however, there
is at least one solution a V with T (a) = b. All
the rest can be found from a by adding solutions x
of the associated homogeneous equations, that is,

Theorem 2. Let V W and W X be linear

transformations. Then
ker(T ) ker(U T )
and
im(U T ) im(U ).
Proof. For the first statement, just note that
T (v) = 0 implies U (T (v)) = 0. For the second
statement, just note that an element of the form
U (T (v) in X is automatically of the form U (w)
where w = T (v).
q.e.d.

Definition 3. The dimensions of the kernel and

image of a transformation T are called the transT (a + x) = b iff T (x) = 0.
formations rank and nullity, and theyre denoted
rank(T ) and nullity(T ), respectively. Since a ma- Geometrically, the solution set is a translate of the
trix represents a transformation, a matrix also has kernel of T , which is a subspace of V , by the vector
a rank and nullity.
a.
1

Example 5. Consider this nonhomogeneous linear

system Ax = b:


 x
 
2 0 1
1
y =
1 1 3
2
z

The dimension theorem. The rank and nullity

of a transformation are related. Specifically, their
sum is the dimension of the domain of the transformation. That equation is sometimes called the
dimension theorem.
In terms of matrices, this connection can be
stated as the rank of a matrix plus its nullity equals
the number of rows of the matrix.
Before we prove the Dimension Theorem, first
well find a characterization of the image of a transformation.

Solve it by row reducing the augmented matrix



2 0 1 1
1 1 3 2
to

1 0 1/2 1/2
0 1 5/2 3/2

Theorem 6. The image of a transformation is

spanned by the image of the any basis of its doThen z can be chosen freely, and x and y deter- main. For T : V W , if = {b1 , b2 , . . . , bn } is a
basis of V , then T () = {T (b1 ), T (b2 ), . . . , T (bn )}
mined from z, that is,
spans the image of T .
1 1 1 1
2z
x
2
2
2
3
3
5
Although T () spans the image, it neednt be a

x = y = 2 2 z = 2 + 52 z
basis
because its vectors neednt be independent.
0
1
z
z
Proof. A vector in the image of T is of the form
T (v) where v V . But is a basis of V ,
so v is a linear combination of the basis vectors
{b1 , . . . , bn }. Therefore, T (v) is same linear combination of the vectors T (b1 ), . . . , T (bn ). Therefore,
every vector in the image of T is a linear combination of vectors in T ().
q.e.d.

This last equation is the parametric equation of a

line in R3 , that is to say, the solution set is a line.
But its not a line through the origin. There is,
however, a line through the origin, namely
1
x
2
x = y = 52 z
1
z

Theorem 7 (Dimension Theorem). If the domain

of a linear transformation is finite dimensional, then
that dimension is the sum of the rank and nullity
of the transformation.

and that line is the solution space of the associated

homogeneous system Ax = 0. Furthermore,
these
1/2

two lines are parallel, and the vector 3/2 shifts

0
the line through the origin to the other line. In
summary, for this example, the solution set for the
nonhomogeneous equation Ax = b is a line in R3
parallel to the solution space for the homogeneous
equation Ax = 0.

Proof. Let T : V W be a linear transformation,

let n be the dimension of V , let r be the rank of T
and k the nullity of T . Well show n = r + k.
Let = {b1 , . . . , bk } be a basis of the kernel
of T . This basis can be extended to a basis =
{b1 , . . . , bk , . . . , bn } of all of V . Well show that
the image of the vectors we appended,

In general, the solution set for the nonhomogeneous equation Ax = b wont be a one-dimensional
C = {T (bk+1 ), . . . , T (bn )}
line. Its dimension will be the nullity of A, and it
will be parallel to the solution space of the associ- is a basis of T (V ). That will show that r = n k
ated homogeneous equation Ax = 0.
as required.
2

First, we need to show that the set C spans the

image T (V ). From the previous theorem, we know
that T () spans T (V ). But all the vectors in T ()
are 0, so they dont help in spanning T (V ). That
leaves C = T () T () to span T (V ).
Next, we need to show that the vectors in C are
linearly independent. Suppose that 0 is a linear
combination of them,

of 0 alone. Thats a 0-dimensional space, so the

nullity of T is 0.
Conversely, suppose that the nullity of T is 0,
that is, its kernel consists only of 0. Well show T is
one-to-one. Suppose that u 6= v. Then u v 6= 0.
Then u v 6= 0 does not lie in the kernel of T
which means that T (u v) 6= 0. But T (u v) =
T (u) T (v), therefore T (u) 6= T (v). Thus, T is
one-to-one.
q.e.d.

ck+1 T (bk+1 ) + + cn T (bn ) = 0

Since the rank plus the nullity of a transformation equals the dimension of its domain, r + k = n,
we have the following corollary.

where the ci s are scalars. Then

T (ck+1 bk+1 + + bn vn ) = 0

Corollary 9. For a transformation T whose domain is finite dimensional, T is one-to-one if and

only if that dimension equals its rank.

Therefore, v = ck+1 bk+1 + + cn bn lies in the

kernel of T . Therefore, v is a linear combination of
the basis vectors , v = c0 b0 + +ck bk . These last
two equations imply that 0 is a linear combination
of the entire basis of V ,

Characterization of isomorphisms. If two

vector spaces V and W are isomorphic, their propc0 b0 + + ck bk ck+1 bk+1 cn bn = 0.
erties with respect to addition and scalar multiplication will be the same. A subset of V will span V
Therefore, all the coefficients ci are 0. Therefore, if and only if its image in W spans W . Its indepenthe vectors in C are linearly independent.
dent in V if and only if its image is independent in
Thus, C is a basis of the image of T .
q.e.d. W . Therefore, an isomorphism sends any basis of
There are several results that follow from this the first to a basis of the second. And that means
Dimension Theorem.
Theorem 10. Isomorphic vector spaces have the
same dimension.
Characterization of one-to-one transformations. Recall that a function f is said to be
Well be interested in which linear transformaa one-to-one function if whenever x 6= y then tions are isomorphisms. The Dimension Theorem
f (x) 6= f (y). That applies to transformations as gives us some criteria for that.
well. T : V W is one-to-one when u 6= v implies
T (u) 6= T (v). An alternate term for one-to-one is Theorem 11. Let T : V W be a linear transinjective, and yet another term thats often used for formation be a transformation between two vector
spaces of the same dimension. Then the following
transformations is monomorphism.
Which transformations are one-to-one can be de- statements are equivalent.
termined by their kernels.
(1) T is an isomorphism.
(2) T is one-to-one.
Theorem 8. A transformation is one-to-one if and
(3) The nullity of T is 0, that is, its kernel is
only if its kernel is trivial, that is, its nullity is 0.
trivial.
(4) T is onto, that is, the image of T is W .
Proof. Let T : V W . Suppose that T is one-to(5) The rank of T is the same as the dimension
one. Then since T (0) = 0, therefore T can send no
of
the vector spaces.
other vector to 0. Thus, the kernel of T consists
3

Proof. (2) and (3) are equivalent by a previous theorem. (4) and (5) are equivalent by the definition of
rank. Since n = r + k, (3) is equivalent to (5). (1)
is equivalent to the conjunction of (2) and (3), but
theyre equivalent to each other, so theyre equivalent to (1).
q.e.d.
Math 130 Home Page at
http://math.clarku.edu/~djoyce/ma130/