1

MATHEMATICS FOR DATA SCIENCE II

WEEK 5

TOPICS TO BE COVERED IN WEEK 5

 Null spaces
 Nullity
 Rank-Nullity Theorem
 Linear Transformation

Null space of a matrix

Let A be an m × n matrix. The subspace {x ∈Rn : Ax = 0} is called the null space of the matrix
A. This is nothing but the solution space of the homogeneous system of linear equations Ax
= 0.

Nullity of a matrix

The dimension of the null space of A is the nullity of the matrix A.

The rank-nullity theorem

Let A be an m × n matrix. Then rank(A)+nullity(A) = n.

Examples:

Let nullity of the matrix Am×n be 2.

Then the rank of A can be got using the rank-nullity theorem.

Rank(A) + nullity(A) = n.

Thus rank(A) = 2.

Exercises:

1) Choose the correct set of options from the following.

• Option 1: The nullity of a non-zero scalar matrix of order 3 must be 3.

• Option 2: The nullity of a non-zero scalar matrix of order 3 must be 0.
• Option 3: The nullity of a non-zero diagonal matrix of order 3 must be 3.
• Option 4: The nullity of a non-zero diagonal matrix of order 3 can be at most 2.
 2

2) Which of the following options are correct for a square matrix A of order n × n, where n
is any natural number?
• Option 1: If the determinant is non-zero, then the nullity of A must be 0.
• Option 2: If the determinant is non-zero, then the nullity of A may be non-zero.
• Option 3: If the nullity of A is non-zero, then the determinant of A must be 0.
• Option 4: If the nullity of A is non-zero, then the determinant of A may be non-zero.

3) If A is a 3 × 4 matrix, then which of the following options are true?

• Option 1: rank(A) must be less than or equal to 3.
• Option 2: nullity(A) must be greater than or equal to 1.
• Option 3: If A has 2 columns which are non-zero and not multiples of each other,
while the remaining columns are linear combinations of these 2 columns, then
nullity(A) = 2.
• Option 4: If A has 2 columns which are non-zero and not multiples of each other,
while the remaining columns are linear combinations of these 2 columns, then
nullity(A) = 1
 3

Notice what A has done: it took a vector in R3 and transformed it into a vector in R2.
 4

Linear Transformation

A function T : V → W between two vector spaces V and W is said to be a linear

transformation if for any two vectors v1 and v2 in the vector space V and for any c ∈ R
(scalar) the following conditions hold:

• T (v1 + v2) = T (v1) + T (v2)

• T (cv1) = cT (v1)

Examples:

1) Consider a mapping T : R2 → R2 defined as T (x, y) = (2x+y,x- y). Is T a linear

transformation?

Any linear transformation f:Rn Rm can be defined as follows:

Fact: For a linear transformation T : V → W , where V , W are vector spaces, T (0) = 0 i.e.,
Image of the zero vector of V is the zero vector of W .
 5

2) Let us choose the standard basis {(1, 0), (0, 1)} of R 2. Define the linear transformation
as follows:
T:R2→R2
T ((1, 0)) = (2, 0)
T ((0, 1)) = (0, 1)
Find the explicit definition of T.

If we choose a different basis for R 2 , then we will get different linear transformation. Let
us choose {(1, 0), (1, 1)} to be a basis for R 2. Define the linear transformation as we have
defined earlier: f : R2 → R2 f((1, 0)) = (2, 0) f((1, 1)) = (1, 1)
 6

Observation: Let T : V → W be a linear transformation. It is enough to know the image of

the basis elements of V to get the explicit definition of T.
 7

3) Let V be a vector space, and let x ≠0 be a fixed vector in V. Is the translation function
f : V V defined by f (v) = v + x is a linear transformation?

T : R2 → R2
T(x, y) = (x+1, y-2)

4) Choose the set of correct options.

Option 1: Let u = (3, 1, 0), v = (0, 1, 7), and w = (3, 0, −7).
There is a linear transformation T : R3 → R3 such that T (u) = T (v) = (0, 0, 0) and that T
(w) = (5, 1, 0).
Option 2: Let u = (2, 1, 0), v = (1, 0, 1), and w = (3, 5, 6). There is a linear transformation
T : R3 → R2 such that T (u) = (1, 0), T (v) = (0, 1) and that T (w) = (5, 6).
Option 3: Let u = (1, 0, 0), v = (1, 0, 1), and w = (0, 1, 0). There is a linear transformation
T : R3 → R3 such that T (u) = (0, 0, 0), T (v) = (0, 0, 0), T (w) = (0, 0, 0), and T (1, 4, 2) = (2,
4, 1).
Option 4: Let u = (1, 0), and v = (0, 1). There is a linear transformation T : R2 → R2 such
that T (u) = (π, 1), and T (v) = (1, e).
 8

5) Suppose V is any vector space. Define a linear transformation T: V V?

If T:Rn→Rm is a linear transformation if and only if there exists a m×n matrix A such
that T(v)=Av for all v∈Rn.

6) Consider the following linear transformation T defined as follows:

T : R2 → R2
a) T(x, y) = (x, -y) b) T(x, y) = (-x, y)
 9

7) Consider the mapping g: Rn → Rn given by scalar multiplication by k, where k ∈ R; that

is, g(v) =kv, for v ∈Rn. If |k| > 1, g represents a dilation (lengthening) of the vectors in Rn;
 10

if |k| < 1, g represents a contraction (shrinking).

 11

8)

9) Consider the mapping h: R3 → R3 given by h([a1 , a2 , a3]) = [a1 , a2 , 0]. This mapping
takes each vector in R3 to a corresponding vector in the xy-plane.

10)
 12

Recall: A function is called injective if there are no two elements from the domain which
map to a same image.
A function is called surjective if every element of the codomain of the function has a
pre-image.

Similarly,
1. a linear transformation T : V → W , where V , W are vector spaces, is called a
monomorphism, if T is an injective map from V to W , i.e., T (v1) = T (v2) ⇒ v1 = v2.
2. a linear transformation T : V → W , where V , W are vector spaces, is called an
epimorphism, if T is a surjective map from V to W , i.e., for every w ∈ W there exists v
∈ V such that T (v) = w.
3. a linear transformation T : V → W is isomorphism if it is both injective
(monomorphism) and surjective (epimorphism).
 13

Examples:

11) Consider the following linear transformation T defined as follows:

T : R2 → R2
(a) T (x, y) = (2x, y) (b) T (x, y) = (2x, 0)
Is T an isomorphism?
 14

12) Choose the correct set of options:

⃝ Option 1: T is a one to one linear transformation if and only if there does
not exist any v≠0 ∈ V so that T (v) = 0.
⃝ Option 2: If T : R2 → R2 is a surjective linear transformation, then it cannot
be injective.
⃝ Option 3: If T : R → R is a linear transformation which is not injective, then
T must be 0, i,e., T (v) = 0 for all v ∈ R.
⃝ Option 4: If there exists some non-zero vector v ∈V , such that T (v) = 0,
for a linear transformation T : V → W , then T cannot be an isomorphism.

13) Let S : V1 → V2 and T : V2 → V3 be two linear transformations. Let us define

T ◦ S : V1 → V3 by T ◦ S(v) = T (S(v)). Choose the correct set of options.
⃝ Option 1: If T ◦ S is injective, then T must be injective.
⃝ Option 2: If T ◦ S is injective, then S must be injective.
⃝ Option 3: If T ◦ S is surjective, then T must be surjective.
⃝ Option 4: If T ◦ S is surjective, then S must be surjective.
 15

Option 1: T is injective if and only if the system of linear equations Av = 0

has a unique solution.

 16

⃝ Option 2: T is injective if and only if rank(A) = 3.

⃝ Option 3: If the system of linear equations Av = b for any b ∈ R3, has a

solution, then T must be surjective.

⃝ Option 4: If T is not surjective, then there exists some b ∈ R3, such that the

system of linear equations Av = b has no solution.

17
18
19
20
21