Module 4

(Spanning Set)
 Linear Combination

Linear combination

Definition:
A linear combination of a list 𝑣1 , 𝑣2 , … , 𝑣𝑘 of vectors in 𝑉
is a vector of the form
𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑘 𝑣𝑘 ,
where 𝑎1 , 𝑎2 , … , 𝑎𝑘 ∈ 𝐹.
Example-1
(17, −4, 2) is a linear combination of 2, 1, −3 , (1, −2, 4) because
17, −4, 2 = 6 2, 1, −3 + 5 1, −2, 4 .
Example-1
(17, −4, 2) is a linear combination of 2, 1, −3 , (1, −2, 4) because
17, −4, 2 = 6 2, 1, −3 + 5 1, −2, 4 .

Example-2
(17, −4, 5) is not a linear combination of 2, 1, −3 , (1, −2, 4) because
there do not exist numbers 𝑎1 , 𝑎2 ∈ 𝐹 such that
17, −4, 5 = 𝑎1 2, 1, −3 + 𝑎2 1, −2, 4 .
 Spanning Set

Span
Definition: If 𝑆 = {𝑣1 , 𝑣2 , … , 𝑣𝑘 } is a set of vectors in a
vector space 𝑉, then the set of all linear combinations of
𝑣1 , 𝑣2 , … , 𝑣𝑘 is called the span of 𝑣1 , 𝑣2 , … , 𝑣𝑘 and is
denoted by span(𝑣1 , 𝑣2 , … , 𝑣𝑘 ) or span(𝑆).

• If 𝑉 = span(𝑆), then 𝑆 is called a spanning set for 𝑉

and 𝑉 is said to be spanned by 𝑆.
• In other words, span 𝑣1 , 𝑣2 , … , 𝑣𝑘 =
{𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑘 𝑣𝑘 : 𝑎1 , 𝑎2 , … , 𝑎𝑘 ∈ 𝐹}

Note: The span of the empty list ( ) is defined to be {0}.

Example-3
1. { 1, 0 , 0, 1 } span ℝ2 .
Example-3
1. { 1, 0 , 0, 1 } span ℝ2 .

2. { 1, 0, 0 , 0, 1, 0 , 0, 0, 1 } span ℝ3 .
Example-3
1. { 1, 0 , 0, 1 } span ℝ2 .

2. { 1, 0, 0 , 0, 1, 0 , 0, 0, 1 } span ℝ3 .

3. The polynomials 1, 𝑥, 𝑥 2 span 𝒫2 .

Example-3
1. { 1, 0 , 0, 1 } span ℝ2 .

2. { 1, 0, 0 , 0, 1, 0 , 0, 0, 1 } span ℝ3 .

3. The polynomials 1, 𝑥, 𝑥 2 span 𝒫2 .

1 0 0 1 0 0 0 0
4. , , , span 𝑀22 , vector space of
0 0 0 0 1 0 0 1
all 2 × 2 matrices.

• Find one spanning set of 𝒫(F), set of all polynomials.

 Theorems

Theorem 1:
Let 𝑣1 , 𝑣2 , … , 𝑣𝑘 be vectors in a vector space 𝑉. Then
a. span 𝑣1 , 𝑣2 , … , 𝑣𝑘 is a subspace of 𝑉.
b. span 𝑣1 , 𝑣2 , … , 𝑣𝑘 is the smallest subspace
of 𝑉 that contains 𝑣1 , 𝑣2 , … , 𝑣𝑘 .
Theorem 1:
Let 𝑣1 , 𝑣2 , … , 𝑣𝑘 be vectors in a vector space 𝑉. Then
a. span 𝑣1 , 𝑣2 , … , 𝑣𝑘 is a subspace of 𝑉.

Proof:
span(𝑣₁, 𝑣₂, …, 𝑣ₖ) is indeed a subspace of 𝑉.

The span of a set of vectors is the set of all possible linear

combinations of those vectors.

Since it is closed under addition and scalar multiplication, it

satisfies the properties of a subspace.
Theorem 1:
Let 𝑣1 , 𝑣2 , … , 𝑣𝑘 be vectors in a vector space 𝑉. Then
b. span 𝑣1 , 𝑣2 , … , 𝑣𝑘 is the smallest subspace of 𝑉
that contains 𝑣1 , 𝑣2 , … , 𝑣𝑘 .

Proof:
span(𝑣₁, 𝑣₂, …, 𝑣ₖ) is the smallest subspace of 𝑉 that contains
𝑣₁, 𝑣₂, …,𝑣ₖ. This is because any subspace that contains 𝑣₁,
𝑣₂, …, 𝑣ₖ must contain all possible linear combinations of
these vectors, and that set of linear combinations is the
span(𝑣₁, 𝑣₂, …, 𝑣ₖ).

Therefore, it is the smallest that includes the given vectors.

Theorem 2: Spanning set of a vector space is not unique.

Proof: Let 𝑆 = {𝑣1 , 𝑣2 , … , 𝑣𝑘 } is a spanning set of 𝑉.

Then, any element 𝑣 can be written as linear combination
of 𝑆, i.e., 𝑣 = 𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑘 𝑣𝑘 ,
where 𝑎1 , 𝑎2 , … , 𝑎𝑘 ∈ 𝐹.
Now, consider 𝑆1 = {𝑎𝑣1 , 𝑣2 , … , 𝑣𝑘 }, 𝑎 ≠ 0.
Then one can write the element 𝑣 as linear combination
of 𝑆1 ,
i.e., 𝑣 = 𝑏1 𝑎𝑣1 + 𝑏2 𝑣2 + ⋯ + 𝑏𝑘 𝑣𝑘 ,
where 𝑏1 , 𝑏2 , … , 𝑏𝑘 ∈ 𝐹.
Hence, 𝑆1 is also a spanning set. So, spanning set is not
unique.
Theorem 3: Let 𝑣1 , 𝑣2 , … , 𝑣𝑘 be vectors in ℝk . Then
span 𝑣1 , 𝑣2 , … , 𝑣𝑘 is a subspace of ℝk .
Proof:
The span is defined as the set of all possible linear
combinations of these vectors, and it satisfies the properties
of a subspace:
Contains the zero vector: The span always contains the zero
vector because
𝟎 = 0𝑣1 + 0𝑣2 + ⋯ + 0𝑣𝑘 .
Closed under addition: If 𝒗 and 𝒘 are in the span, then any
linear combination
𝑎𝒗 + 𝑏𝒘, 𝑎, 𝑏 ∈ 𝐹 is also in the span.
Closed under scalar multiplication: If 𝒗 is in the span, then
any scalar multiple 𝑐𝒗 is also in the span.
Therefore, span(𝑣₁, 𝑣₂, … , 𝑣𝑘 ) is a subspace of ℝk .
Theorem 4: Let 𝑣1 , 𝑣2 , … , 𝑣𝑘 be vectors in ℝk . Then {𝑣1 , 𝑣2 , … , 𝑣𝑘 }
spans ℝk if and only if, for the matrix 𝐴 = [𝑣1 , 𝑣2 , … , 𝑣𝑘 ], the
linear system 𝐴𝑐 = 𝑣 is consistent for every 𝑣 in ℝk .
Row and column spaces of a matrix

Definition:
Let 𝐴 be an 𝑚 × 𝑛 matrix.
n
• The row space of 𝐴 is the subspace row(𝐴) of ℝ spanned
by the rows of 𝐴.
m
• The column space of 𝐴 is the subspace col(𝐴) of ℝ
spanned by the columns of 𝐴.
 1 −5 2
Example-1: Consider the matrix 𝐴 = . Then,
6 0 7

Row space = subspace of ℝ3 spanned by { 1, −5, 2 , (6, 0, 7)}.

Column space = subspace of ℝ2 spanned by 1, 6 , −5, 0 , 2, 7 ,

= subspace of ℝ2 spanned by 1, 6 , −5, 0 .
since,
7 1
1, 6 − −5, 0 = 2, 7 .
6 6
Row equivalence
Definition: Two matrices 𝐴, 𝐵 are said to be row equivalent if
one can be changed to the other by a sequence of elementary
row operations.

1 0 0 1 0 0
Example-2: The matrices A = and B =
0 1 1 1 1 1
are row equivalent as matrix 𝐴 can be obtained by applying the
row operation 𝑅2′ = 𝑅2 − 𝑅1 on the matrix 𝐵.
 1 1
Example-3: The column space of consists of all linear
0 0
combinations of the form
1 1
𝑐1 +𝑐2 .
0 0
1 𝑥
It is the span of . If we identify the column vectors 𝑦 with
0
points (x, y), we may picture this span as follows:

1 1
Question-4: Can you draw the row space for the matrix ?
0 0
Theorem:
Let 𝐵 be any matrix that is row equivalent to a matrix 𝐴. Then
row(𝐵) = row(𝐴).
 Module 4
(Linear Dependence and Independence )
 Linear Independence of Vectors

Linear Independent:
A list of vectors 𝑣1 , 𝑣2 , … , 𝑣𝑘 in 𝑉 is called linearly
independent if the only choice of 𝑎1 , 𝑎2 , … , 𝑎𝑘 ∈ 𝐹 that
makes 𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑘 𝑣𝑘 equal 0 is 𝑎1 = 𝑎2 =
⋯ = 𝑎𝑘 = 0.

Note: The empty list ( ) is also declared to be linearly

independent.
Example
a. A list of one vector 𝑣 ∈ 𝑉 is linearly independent if
and only if 𝑣 ≠ 0.
Example
a. A list of one vector 𝑣 ∈ 𝑉 is linearly independent if
and only if 𝑣 ≠ 0.

b. A list of two vectors in 𝑉 is linearly independent if and

only if neither vector is a scalar multiple of the other.
Example
a. A list of one vector 𝑣 ∈ 𝑉 is linearly independent if
and only if 𝑣 ≠ 0.

b. A list of two vectors in 𝑉 is linearly independent if and

only if neither vector is a scalar multiple of the other.

c. { 1, 0, 0, 0 , 0, 1, 0, 0 , (0, 0, 1, 0)} is linearly

independent in ℝ4 .
Example
a. A list of one vector 𝑣 ∈ 𝑉 is linearly independent if
and only if 𝑣 ≠ 0.

b. A list of two vectors in 𝑉 is linearly independent if and

only if neither vector is a scalar multiple of the other.

c. { 1, 0, 0, 0 , 0, 1, 0, 0 , (0, 0, 1, 0)} is linearly

independent in ℝ4 .

d. The list 1, 𝑥, 𝑥 2 , … , 𝑥 𝑘 is linearly independent in

𝒫𝑘 for each non-negative integer 𝑘.
 Linear Dependence of Vectors

Linear Dependent
A list of vectors in 𝑉 is called linearly dependent if it is not
linearly independent. In other words, a list 𝑣1 , 𝑣2 , … , 𝑣𝑘 in
𝑉 is linearly dependent if there exist 𝑎1 , 𝑎2 , … , 𝑎𝑘 ∈ 𝐹 ,
not all 0, such that
𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑘 𝑣𝑘 = 0.
Example
a. 2,3, 1 , 1, −1,2 , 7, 3, 8 is linearly dependent in ℝ3
because
2 2,3, 1 , +3 1, −1,2 , + −1 7, 3, 8 = 0, 0, 0 .
Example
a. 2,3, 1 , 1, −1,2 , 7, 3, 8 is linearly dependent in ℝ3
because
2 2,3, 1 , +3 1, −1,2 , + −1 7, 3, 8 = 0, 0, 0 .

b. The list 2, 3, 1 , 1, −1, 2 , (7, 3, 𝑐) is linearly dependent

in ℝ3 if and only if 𝑐 = 8. (verify)
Example
a. 2,3, 1 , 1, −1,2 , 7, 3, 8 is linearly dependent in ℝ3
because
2 2,3, 1 , +3 1, −1,2 , + −1 7, 3, 8 = 0, 0, 0 .

b. The list 2, 3, 1 , 1, −1, 2 , (7, 3, 𝑐) is linearly dependent

in ℝ3 if and only if 𝑐 = 8. (verify)

c. Let 𝑢 = 1, 1, 0 , 𝑣 = 1, 3, 2 , 𝑤 = (4,9,5). Then 𝑢, 𝑣, 𝑤

are linearly dependent.
1. Find a number 𝑡 such that 3, 1, 4 , 2, −3, 5 , 5,9, 𝑡
is not linearly independent in ℝ3 .

2. Suppose 𝑣1 , 𝑣2 , 𝑣3 , 𝑣4 is linearly independent in 𝑉.

Prove that the list
𝑣1 − 𝑣2 , 𝑣2 − 𝑣3 , 𝑣3 − 𝑣4 , 𝑣4
is also linearly independent.

3. Prove or give a counter example:

If 𝑣1 , 𝑣2 , … , 𝑣𝑘 is a linearly independent list of
vectors in 𝑉, then 5𝑣1 − 4𝑣2 , 𝑣2 , … , 𝑣𝑘 is linearly
independent.
 Some Remarks

Remark 1:
Suppose 𝟎 is one of the vectors 𝑣1 , 𝑣2 , … , 𝑣𝑘 , say 𝑣1 =
0. Then the vectors must be linearly dependent.

Remark 2:
Suppose 𝑣 is a nonzero vector. Then 𝑣, by itself, is
linearly independent.
Remark 3:
Suppose two of the vectors 𝑣1 , 𝑣2 , … , 𝑣𝑘 are equal or one
is a scalar multiple of the other, say 𝑣1 = 𝑘𝑣2 . Then the
vectors must be linearly dependent.

Remark 4:
Two vectors 𝑣1 and 𝑣2 are linearly independent if and only
if one of them is a multiple of other.
Remark 5:
If the set {𝑣1 , 𝑣2 , … , 𝑣𝑘 } is linearly independent, then any
rearrangement of the vectors{𝑣𝑖1 , 𝑣𝑖2 , … , 𝑣𝑖𝑘 } is also
linearly independent.

Remark 6:
If a set 𝑆 of vectors is linearly independent, then any
subset of 𝑆 is linearly independent. Alternatively, if 𝑆
contains a linearly dependent subset, then 𝑆 is linearly
dependent
 Module 4
(Basis and Dimension)
 Basis and Dimension

Basis:
A set 𝑆 = {𝑢1 , 𝑢2 , … , 𝑢𝑛 } of vectors is a basis of a 𝑉 if it
has the following two properties:
a). 𝑆 is linearly independent
b). 𝑆 spans 𝑉.

Note: Basis of a vector space is not unique.

Example
1. The list 1, 0, 0, … , 0 , 0, 1, 0, … , 0 , … , (0, 0, 0, … , 1)
is a basis of ℝn , called the standard basis of ℝn .
Example
1. The list 1, 0, 0, … , 0 , 0, 1, 0, … , 0 , … , (0, 0, 0, … , 1)
is a basis of ℝn , called the standard basis of ℝn .

2. The list 1, 2 , (3, 5) is a basis of ℝ2 .

Example
1. The list 1, 0, 0, … , 0 , 0, 1, 0, … , 0 , … , (0, 0, 0, … , 1)
is a basis of ℝn , called the standard basis of ℝn .

2. The list 1, 2 , (3, 5) is a basis of ℝ2 .

3. The list 1, 2, −4 , (7, −5, 6) is linearly independent

in ℝ3 but is not a basis of ℝ3 because it does not
span ℝ3 .
Example
1. The list 1, 0, 0, … , 0 , 0, 1, 0, … , 0 , … , (0, 0, 0, … , 1)
is a basis of ℝn , called the standard basis of ℝn .

2. The list 1, 2 , (3, 5) is a basis of ℝ2 .

3. The list 1, 2, −4 , (7, −5, 6) is linearly independent

in ℝ3 but is not a basis of ℝ3 because it does not
span ℝ3 .

4. The list 1, 2 , 3, 5 , (4, 13) spans ℝ2 but is not a

basis of ℝ2 because it is not linearly independent.
Example
1. The list 1, 0, 0, … , 0 , 0, 1, 0, … , 0 , … , (0, 0, 0, … , 1)
is a basis of ℝn , called the standard basis of ℝn .

2. The list 1, 2 , (3, 5) is a basis of ℝ2 .

3. The list 1, 2, −4 , (7, −5, 6) is linearly independent

in ℝ3 but is not a basis of ℝ3 because it does
not span ℝ3 .

4. The list 1, 2 , 3, 5 , (4, 13) spans ℝ2 but is not a

basis of ℝ2 because it is not linearly independent.

5. The list 1, 𝑥, 𝑥 2 , … , 𝑥 𝑘 is is a basis of 𝒫𝑘 for each

non-negative integer 𝑘.
Dimension:
If S is a subspace of ℝn , then the number of vectors in a
basis for 𝑆 is called the dimension of 𝑆, denoted dim 𝑆.

Note: Dimension is unique.

Finite Dimensional Vector Space:
A vector space 𝑉 is said to be of finite dimension 𝑛 or
n-dimensional, written
dim 𝑉 = 𝑛,
if 𝑉 has a basis with 𝑛 elements.

• The vector space 0 is defined to have dimension 0.

Infinite Dimensional Vector Space:

Suppose a vector space 𝑉 does not have a finite basis.
Then V is said to be of infinite dimension or to be infinite
dimensional.
Example
1. The list 1, 0, 0, … , 0 , 0, 1, 0, … , 0 , … , (0, 0, 0, … , 1) is a basis of
ℝn , called the standard basis of ℝn . Dimension = 𝑛.
Example
1. The list 1, 0, 0, … , 0 , 0, 1, 0, … , 0 , … , (0, 0, 0, … , 1) is a basis of
ℝn , called the standard basis of ℝn . Dimension = 𝑛.

2. The list 1, 2 , (3, 5) is a basis of ℝ2 . Dimension = 2.

Example
1. The list 1, 0, 0, … , 0 , 0, 1, 0, … , 0 , … , (0, 0, 0, … , 1) is a basis of
ℝn , called the standard basis of ℝn . Dimension = 𝑛.

2. The list 1, 2 , (3, 5) is a basis of ℝ2 . Dimension = 2.

3. The list 1, 𝑥, 𝑥 2 , … , 𝑥 𝑘 is is a basis of 𝒫𝑘 for each non-negative

integer 𝑘. Dimension = 𝑘 + 1.
Example
1. The list 1, 0, 0, … , 0 , 0, 1, 0, … , 0 , … , (0, 0, 0, … , 1) is a basis of
ℝn , called the standard basis of ℝn . Dimension = 𝑛.

2. The list 1, 2 , (3, 5) is a basis of ℝ2 . Dimension = 2.

3. The list 1, 𝑥, 𝑥 2 , … , 𝑥 𝑘 is is a basis of 𝒫𝑘 for each non-negative

integer 𝑘. Dimension = 𝑘 + 1.

1 1 3 1 6
2 −1 0 1 −1
4. Consider the matrix A = . The reduced
−3 2 1 −2 1
4 1 6 1 3
1 0 1 0 −1
0 1 2 0 3
row echelon form of 𝐴 is . The basis of the row
0 0 0 1 4
0 0 0 0 0
space is { 1, 0, 1, 0, −1 , 0, 1, 2, 0, 3 , (0, 0, 0, 1, 4)}.
Dimension = 3.
 Some Theorems on Basis and Dimension

Theorem:
A list (𝑣1 , 𝑣2 , … , 𝑣𝑛 )of vectors in 𝑉 is a basis of 𝑉 if and only if every
𝑣 ∈ 𝑉 can be written uniquely in the form
𝑣 = 𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑛 𝑣𝑛 , …………………..(𝑖)

where 𝑎1 , 𝑎2 ,… 𝑎𝑛 ∈ F.
Theorem:
A list (𝑣1 , 𝑣2 , … , 𝑣𝑛 )of vectors in 𝑉 is a basis of 𝑉 if and only if every
𝑣 ∈ 𝑉 can be written uniquely in the form
𝑣 = 𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑛 𝑣𝑛 , …………………..(𝑖)

where 𝑎1 , 𝑎2 ,… 𝑎𝑛 ∈ F.

Proof: First suppose that (𝑣1 , 𝑣2 , … , 𝑣𝑛 ) is a basis of 𝑉. Let 𝑣 ∈ 𝑉.

Because (𝑣1 , 𝑣2 , … , 𝑣𝑛 ) spans 𝑉, there exist 𝑎1 , 𝑎2 ,… 𝑎𝑛 ∈ F such
that
𝑣 = 𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑛 𝑣𝑛 holds.
Theorem:
A list (𝑣1 , 𝑣2 , … , 𝑣𝑛 )of vectors in 𝑉 is a basis of 𝑉 if and only if every
𝑣 ∈ 𝑉 can be written uniquely in the form
𝑣 = 𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑛 𝑣𝑛 , …………………..(𝑖)

where 𝑎1 , 𝑎2 ,… 𝑎𝑛 ∈ F.

Proof: First suppose that (𝑣1 , 𝑣2 , … , 𝑣𝑛 ) is a basis of 𝑉. Let 𝑣 ∈ 𝑉.

Because (𝑣1 , 𝑣2 , … , 𝑣𝑛 ) spans 𝑉, there exist 𝑎1 , 𝑎2 ,… 𝑎𝑛 ∈ F such
that
𝑣 = 𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑛 𝑣𝑛 holds.

To show that the above representation is unique, suppose that

𝑏1 , 𝑏2 ,… 𝑏𝑛 are scalars so that
𝑣 = 𝑏1 𝑣1 + 𝑏2 𝑣2 + ⋯ + 𝑏𝑛 𝑣𝑛 , ……………………..(𝑖𝑖)
Subtracting (𝑖) and (𝑖𝑖), we get

0 = (𝑏1 −𝑎1 )𝑣1 + (𝑏2 −𝑎2 )𝑣2 + ⋯ + (𝑏𝑛 −𝑎𝑛 )𝑣𝑛

⇒ 𝑏𝑖 −𝑎𝑖 = 0, (because (𝑣1 , 𝑣2 , … , 𝑣𝑛 ) is linearly independent)
⇒ 𝑎1 = 𝑏1 , 𝑎2 = 𝑏2 , … , 𝑎𝑛 = 𝑏𝑛
This completes the proof of uniqueness.
Subtracting (𝑖) and (𝑖𝑖), we get

0 = (𝑏1 −𝑎1 )𝑣1 + (𝑏2 −𝑎2 )𝑣2 + ⋯ + (𝑏𝑛 −𝑎𝑛 )𝑣𝑛

⇒ 𝑏𝑖 −𝑎𝑖 = 0, (because (𝑣1 , 𝑣2 , … , 𝑣𝑛 ) is linearly independent)
⇒ 𝑎1 = 𝑏1 , 𝑎2 = 𝑏2 , … , 𝑎𝑛 = 𝑏𝑛
This completes the proof of uniqueness.

For the other direction, suppose that every 𝑣 ∈ 𝑉 can be written uniquely
in the form
𝑣 = 𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑛 𝑣𝑛 .
⇒ (𝑣1 , 𝑣2 , … , 𝑣𝑛 ) spans 𝑉.
Subtracting (𝑖) and (𝑖𝑖), we get

0 = (𝑏1 −𝑎1 )𝑣1 + (𝑏2 −𝑎2 )𝑣2 + ⋯ + (𝑏𝑛 −𝑎𝑛 )𝑣𝑛

⇒ 𝑏𝑖 −𝑎𝑖 = 0, (because (𝑣1 , 𝑣2 … , 𝑣𝑛 ) is linearly independent)
⇒ 𝑎1 = 𝑏1 , 𝑎2 = 𝑏2 , … , 𝑎𝑛 = 𝑏𝑛
This completes the proof of uniqueness.

For the other direction, suppose that every 𝑣 ∈ 𝑉 can be written uniquely
in the form
𝑣 = 𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑛 𝑣𝑛 .
⇒ (𝑣1 , 𝑣2 , … , 𝑣𝑛 ) spans 𝑉.
To show that (𝑣1 , 𝑣2 … , 𝑣𝑛 ) is linearly independent, suppose that 𝑎1 ,
𝑎2 ,… 𝑎𝑛 ∈ F such that
0 = 𝑎1 𝑣1 + 𝑎2 𝑣2 + ⋯ + 𝑎𝑛 𝑣𝑛 .
The uniqueness of the representation (with v = 0) implies that
𝑎1 = 𝑎2 =… =𝑎𝑛 = 0.
Thus, (𝑣1 , 𝑣2 , … , 𝑣𝑛 ) is linearly independent and hence is a basis of 𝑉.
Theorem:
Every spanning list in a vector space can be reduced to a basis of the
vector space.
Theorem:
Every spanning list in a vector space can be reduced to a basis of the
vector space.

Proof: Suppose (𝑣1 , 𝑣2 , … , 𝑣𝑛 ) spans 𝑉.

We want to remove some of the vectors from (𝑣1 , 𝑣2 , … , 𝑣𝑛 ) so that
the remaining vectors form a basis of 𝑉.

Follow the multistep process as below.

• Start with 𝐵 = (𝑣1 , 𝑣2 , … , 𝑣𝑛 )

• Step 1: If 𝑣1 = 0, delete 𝑣1 from 𝐵. If 𝑣1 ≠ 0, leave 𝐵 unchanged.

• Step 𝒋: If 𝑣𝑗 is in span(𝑣1 , 𝑣2 , … , 𝑣𝑗−1 ), delete 𝑣𝑗 from 𝐵. If 𝑣𝑗 is not

in span(𝑣1 , 𝑣2 , … , 𝑣𝑗−1 ), leave 𝐵 unchanged.
• Stop the process after step n, getting a list 𝐵.

• This list 𝐵 spans 𝑉 because our original list spanned 𝐵 and we have
discarded only vectors that were already in the span of the
previous vectors.

• The process insures that no vector in B is in the span of the

previous ones.

• Thus B is linearly independent. Hence B is a basis of V.

Theorem:
Every linearly independent list of vectors in a finite-dimensional
vector space can be extended to a basis of the vector space.
Theorem:
Every linearly independent list of vectors in a finite-dimensional
vector space can be extended to a basis of the vector space.

Proof: Suppose 𝑉 is finite dimensional and (𝑣1 , 𝑣2 , … , 𝑣𝑚 ) is linearly

independent in 𝑉. We want to extend (𝑣1 , 𝑣2 , … , 𝑣𝑚 ) to a basis of 𝑉.
Follow the multistep process as below.

• First we let (𝑤1 , 𝑤2 … , 𝑤𝑛 ) be any list of vectors in 𝑉 that spans 𝑉.

• Step 1: If w1 is in the span of (𝑣1 , 𝑣2 , … , 𝑣𝑚 ), let 𝐵 = (𝑣1 , 𝑣2 , … , 𝑣𝑚 ).

If 𝑤1 is not in the span of (𝑣1 , 𝑣2 , … , 𝑣𝑚 ), let 𝐵 = (𝑣1 , 𝑣2 , … , 𝑣𝑚 , 𝑤1 ).

• Step j: If 𝑤𝑗 is in the span of 𝐵, leave 𝐵 unchanged. If 𝑤𝑗 is not in the

span of 𝐵, extend 𝐵 by adjoining 𝑤𝑗 to it.
• After each step, 𝐵 is still linearly.

• After step 𝑛, the span of 𝐵 includes all the 𝑤’𝑠. Thus the 𝐵
obtained after step 𝑛 spans 𝑉 and hence is a basis of 𝑉.
Theorem:
Any two bases of a finite-dimensional vector space have the same
length.
 Theorem:
Any two bases of a finite-dimensional vector space have the same
length.

Proof: Suppose 𝑉 is finite dimensional.

• Let 𝐵1 and 𝐵2 be any two bases of 𝑉.

• Then 𝐵1 is linearly independent in 𝑉 and 𝐵2 spans 𝑉, so the

length of 𝐵1 is at most the length of 𝐵2 .

• Interchanging the roles of 𝐵1 and 𝐵2 , we also see that the length

of 𝐵2 is at most the length of 𝐵1 .

• Thus the length of 𝐵1 must equal the length of 𝐵2 , as desired.

Some More Theorems

Theorem 1: Every finite-dimensional vector space has a basis.

Theorem 2: If 𝑉 is finite-dimensional and 𝑈 is a subspace of 𝑉, then

a. dim 𝑈 ≤ dim 𝑉. b. dim 𝑈 = dim 𝑉 if and only if 𝑈 = 𝑉.

Theorem 3: Suppose 𝑉 is finite-dimensional vector space. Then

every linearly independent list of vectors in 𝑉 with length dim 𝑉 is a
basis of 𝑉.

Theorem 4: Suppose 𝑉 is finite-dimensional vector space. Then

every spanning list of vectors in 𝑉 with length dim 𝑉 is a basis of 𝑉.

Theorem 5: The row and column spaces of a matrix 𝐴 have the

same dimension.
 Module 4
(Sum and Direct Sum of Vector Spaces)
 Sum of Vector Subspaces

Sum: (For two subspaces)

Let 𝑉 be a vector space with subspaces 𝑈 and W. Define the sum of𝑈
and W to be
𝑈+W = {𝑢 + 𝑤: 𝑢 ∈ 𝑈, 𝑤 ∈ 𝑊}.

Sum: (For m subspaces)

Let 𝑉 be a vector space with subspaces 𝑈1 , 𝑈2 , … , 𝑈𝑚 . Define the sum
of 𝑈1 , 𝑈2 , … , 𝑈𝑚 to be
𝑈1 + 𝑈2 + ⋯ + 𝑈𝑚 = {𝑢1 + 𝑢2 + ⋯ + 𝑢𝑚 : 𝑢1 ∈ 𝑈1 , 𝑢2 ∈
𝑈2 , … , 𝑢𝑚 ∈ 𝑈𝑚 }.

Example: Suppose U = 𝑥, 0,0 ∈ ℝ3 : 𝑥 ∈ ℝ and W =

0, 𝑦, 0 ∈ ℝ3 : 𝑦 ∈ ℝ , then
𝑈 + 𝑊 = { 𝑥, 𝑦, 0 : 𝑥, 𝑦 ∈ ℝ}.
Theorem:
If 𝑈1 and 𝑈2 are subspaces of a finite-dimensional vector space, then
dim (𝑈1 +𝑈2 ) =dim 𝑈1 + dim 𝑈2 − dim(𝑈1 ∩ 𝑈2 ).

Example: Let 𝑉 = 𝑀2,2 , the vector space of 2 × 2 matrices. Let 𝑈

consist of those matrices whose second row is zero, and let 𝑊 consist
of those matrices whose second column is zero. Then

𝑎 𝑏 𝑎 0 𝑎 𝑏
𝑈= ,W= and U + W = ,
0 0 𝑐 0 𝑐 0

𝑎 0
U∩W= .
0 0
 Direct Sum of Vector Subspaces

Direct Sum: (For two subspaces)

The vector space 𝑉 is the direct sum of its subspaces 𝑈 and
𝑊, denoted by
𝑉 =𝑈⊕𝑊
if every 𝑣 ∈ 𝑉 can be written in one and only one way as 𝑣 =
𝑢 + 𝑤 where 𝑢 ∈ 𝑈 and 𝑤 ∈ 𝑊.

Direct Sum: (For m subspaces)

Suppose 𝑈1 , 𝑈2 , … , 𝑈𝑚 are subspaces of 𝑉. The sum of 𝑈1 + 𝑈2 +
⋯ + 𝑈𝑚 is called a direct sum if each element of 𝑈1 + 𝑈2 + ⋯ +
𝑈𝑚 can be written in only one way as a sum 𝑢1 + 𝑢2 + ⋯ + 𝑢𝑚 ,
where each 𝑢𝑗 is in 𝑈𝑗 .
If 𝑈1 + 𝑈2 + ⋯ + 𝑈𝑚 is a direct sum, then 𝑈1 ⊕ 𝑈2 ⊕ ⋯ ⊕ 𝑈𝑚
denotes 𝑈1 + 𝑈2 + ⋯ + 𝑈𝑚 with the ⊕ notation serving as an
indication that this is a direct sum.
Theorem
Suppose 𝑈 and 𝑊 are subspaces of 𝑉. Then 𝑈 + 𝑊 is a direct sum if
and only if 𝑈 ∩ 𝑊 = {0}.

Example
Suppose U = 𝑥, 𝑦, 0 ∈ ℝ3 : , 𝑥, 𝑦 ∈ ℝ and W = { 0, 0, 𝑧 ∈ ℝ3 : 𝑧 ∈
ℝ}, then
𝑈 ⊕ 𝑊 = ℝ3 .