3

Vector
spaces

© Copyright Univers idad Europea. Todos los derechos reservados 1

 Vectors
A vector in a mathematical space is an ordered set of numbers that
represent its components in a particular coordinate system.

Vectors can be visualized as “Euclidean vectors” or “geometric

vectors”, represented as arrows, where the length of the arrow
represents the magnitude (size) of the vector, and the direction of the
arrow indicates the direction of the vector.

✓ Physics: physical magnitude with value, direction and way 𝐹Ԧ = 𝑚𝑎Ԧ

✓ Geometry: the directing element of a geometric location on the plane or in

space

✓ Algebra: an algebraic element of a set on which the operations addition and

multiplication by a scalar can be applied.
© Copyright Univers idad Europea. Todos los derechos reservados
 Vectors: important remarks

✓ The coordinates or entries of a vector only tell us how far the vector
stretches in the x- and y-directions; not where it is located in space

Usually, the tail But both these

(back end of the vectors would be
vector) is noted as (4,2),
positioned at the even if one is
origin and that standard
would be called position and the
standard other is
position. displaced.

✓ The order of the coordinates matters: for example (2, 3) ≠ (3, 2)

✓ The zero vector is (0,0) and is denoted as 𝟎
✓ The set of all 2D vectors is denoted as ℝ𝟐 , in 3d is ℝ𝟑 and so on
until n-dimensions, that would be ℝ𝒏

© Copyright Univers idad Europea. Todos los derechos reservados

 Operations with vectors
✓ Addition: Specifically, if 𝒗 = (v1, v2) and 𝒘 = (w1, w2), then 𝒗 + 𝒘 =
(v1 + w1, v2 + w2) is the vector from the tail of 𝒗 to the head of 𝒘

✓ Scalar multiplication: if 𝒗 = (v1, v2) and c is a real number, then

c𝒗 = (cv1, cv2) is the vector that points in the same direction as 𝒗,
but is c times as long

✓ Subtraction and scalar division derive from the previous two.

© Copyright Univers idad Europea. Todos los derechos reservados

 Linear combination

It is said that a vector 𝒘 is linear combination, or that depends

linearly, on a set of vectors 𝒗𝟏 , 𝒗𝟐, ⋯ , 𝒗𝒏 if it can be expressed in
the way:

𝒏
𝒘 = 𝛼1 𝒗𝟏 + 𝛼2 𝒗𝟐 + ⋯ + 𝛼𝑛 𝒗𝒏 = ෍ 𝛼𝑖 𝒗𝒊
𝒊=𝟏

where 𝛼𝑖 ∈ ℝ

© Copyright Univers idad Europea. Todos los derechos reservados

 Linear combination
Example:
Given a vector (7, 14)
How can we know if it can be expressed as a linear combination of
vectors (1, 2) and (1, 3)?

To solve this, we formulate the following equation:

(7, 14) = α (1, 2) + β (1, 3)

The following system should therefore hold:
α+β=7
ቊ
2α + 3β = 14
The type of the system will reveal if there is linear dependence
The augmented matrix of the system reveals
that r(A)=r(A*)=2, so it is consistent
determined system, single solution: β =0,
α=7. Vector (7, 14) = 7 (1, 2) , so it is a linear
combination of (1, 2) only.

© Copyright Univers idad Europea. Todos los derechos reservados

 Linear combination
What is the relation between type of system of equations and linear
combinations?
✓ An undetermined consistent system has infinite solutions, so
there will be an infinite number of linear combinations that give
the same result.
✓ In a determined consistent system, there is a single solution, and
there will be only one unique linear combination that gives the
vector in question.
✓ It the system is inconsistent, the vector cannot be expressed as a
linear combination of the others.
• Consistent system = expression of a vector as linear
combination of others, either uniquely (determined) or not
uniquely (undetermined).
• Inconsistent system = the vector is not a linear combination of
the others.

© Copyright Univers idad Europea. Todos los derechos reservados

 Linear combination

Exercises:

Determine if the vector (1,3,6) linearly depends on vectors 0,1,2 ,

1,1,2 and (3, −5,7).

Discuss if vector (3,6) linearly depends on vectors (1,2) and (2,4)

Study the vector dependence of (1,2, −3) with vectors (0,2, −1) and
(1,0,5)

© Copyright Univers idad Europea. Todos los derechos reservados

 Linear dependence and independence

A set of vectors 𝒗𝟏 , 𝒗𝟐 , ⋯ , 𝒗𝒏 is linearly dependent, if there are n

scalars, not all of them null, such that:

𝒏
𝛼1𝒗𝟏 + 𝛼2 𝒗𝟐 + ⋯ + 𝛼𝒏𝒗𝒏 = ෍ 𝛼𝑖 𝒗𝒊 = 𝟎
𝒊=𝟏

If the relationship is fulfilled only if 𝜶𝟏 = 𝜶𝟐 = 𝜶𝟑 = … = 𝜶𝒏 = 0, the

set of vectors is linearly independent.

© Copyright Univers idad Europea. Todos los derechos reservados

 Linear dependence and independence

Exercises:

Demonstrate that vectors (1,0) and (0,1) are linearly independent.

Demonstrate that vectors (1,0,0), (0,1,0) and (0,0,1) are linearly

independent

Study the vector dependence between (2,-3) and (-4,6).

Study the vector dependence between (-1,2,1), (2,0,1), (0,1,-1) and

(1,-1,0).

© Copyright Univers idad Europea. Todos los derechos reservados

 Linear dependence and independence

Statement 1

A set of vectors is linearly dependent if, and only if, one of its vectors
is a linear combination of the others.

Statement 2

If a set of vectors is linearly independent, so is any subset of it.

Statement 3

If a set of vectors is linearly dependent, so is any set that contains it.

© Copyright Univers idad Europea. Todos los derechos reservados

 Definition of vector space

✓ Let’s build a formal definition for an environment “containing”

vector: the vector space.
✓ It is a set not null of:
✓ mathematical elements (referred to as V),
✓ associated with a set of real* numbers (a field of scalars K),
✓ with two associated operations:
✓ an internal one (addition), so that two elements correspond
unequivocally to one element of the set
✓ an external one (product), so that an element and a scalar
correspond unequivocally to an element of the set.
✓this set has to stick to ten axioms (or rules) for all vector 𝒖,
𝒗 and 𝒘 contained in V and all scalars c and d contained in K.
The triplet (V,+, ·) has a vector space structure.
* Strictly speaking, this is a “Real vector space”. In general, a Vector Space can be
defined over any Field of elements, for example Complex Numbers (called the “scalars”).

© Copyright Univers idad Europea. Todos los derechos reservados

 The axioms of vector space
This ten axioms must hold for all vectors 𝒖 , 𝒗 and 𝒘 belonging
to a vector space 𝑉 and for all scalars c and d belonging to K.

Addition (internal operation)

✓ 𝒖 + 𝒗 is in 𝑉 Closure under addition

✓ 𝒖+𝒗 = 𝒗+𝒖 Conmutative property

✓ 𝒖 + 𝒗 + 𝒘 = 𝒖 + (𝒗 + 𝒘) Associative property

✓ There is a vector (called the zero Additive identity (neutral

vector) 𝟎 in 𝑉 such that 𝒖 + 𝟎 = 𝒖 element)

✓ For each 𝒖 in 𝑉, there is a vector Additive inverse (opposite

− 𝒖 satisfying 𝒖 + (−𝒖) = 𝟎 element)
© Copyright Univers idad Europea. Todos los derechos reservados
 The axioms of vector space
This ten axioms must hold for all vectors 𝒖 , 𝒗 and 𝒘 belonging
to a vector space 𝑉 and for all scalars c and d belonging to K.

Scalar multiplication

✓ 𝑐𝒖 is in 𝑉 Closure under scalar multiplication

✓ 𝑐 𝒖 + 𝒗 = 𝑐𝒖 + 𝑐𝒗 Distributive property (vectorial)

✓ 𝑐 + 𝑑 𝒖 =𝑐𝒖 + 𝑑𝒖 Distributive property (scalar)

✓ 𝑐𝑑 𝒖 =𝑐(𝑑𝒖) Pseudo associative property

✓ 1𝒖 = 𝒖 Scalar identity (neutral element)

© Copyright Univers idad Europea. Todos los derechos reservados

 Examples of vector spaces
✓ Real numbers (ℝ𝟏 )

✓ Matrices of dimension mxn.

✓ Polynomials of degree equal to or less than n

✓ Continuous functions f(x) over an interval [a,b]

✓ Vectors on a plane (ℝ𝟐)

✓ Vectors in space (ℝ𝟑 )

✓ Vectors of n coordinates in an n-dimensional space (ℝ𝒏 )

© Copyright Univers idad Europea. Todos los derechos reservados
 Properties of vector spaces
If V is a vector space on a scalar field K, the following properties
must be true for all 𝑎, 𝑏 ∈ 𝐾 and all 𝑢, 𝑣Ԧ ∈ 𝑉:

✓ 𝐚𝟎 = 𝟎, where 𝟎 is the neutral element of the addition in 𝑉

✓ 0𝒗 = 𝟎, where 0 is the neutral element of the addition in 𝐾

✓ If a𝒗 = 𝟎 then, either a = 0 or 𝒗 = 𝟎

✓ If 𝐚𝒗 = 𝒃𝒗 and 𝒗 ≠ 𝟎, then a = b

✓ If 𝐚𝒗 = 𝒂𝒘 and a ≠ 0, then 𝒗 = 𝒘

✓ −𝐚𝒗 = 𝒂 −𝒗 = −𝒂𝒗

© Copyright Univers idad Europea. Todos los derechos reservados

 Generators of vector spaces (or Spanning Sets)

A set of vectors 𝒗𝟏 , 𝒗𝟐 , ⋯ , 𝒗𝒏 is a generating set of a vector space

V, if every vector of V is a linear combination of them.

In other words, the set of vectors 𝒗𝟏, 𝒗𝟐 , ⋯ , 𝒗𝒏 span V if all vectors

of V are linear combinations of the original set.

Examples:

Vectors (1,0) (0,1) (1,1) are a generator set of ℝ2 , since any vector
(x,y) is a linear combination of them.

Vectors (1,0,0), (0,1,0) and (0,0,1) and (1,1,2) are a generator set of
ℝ3 , since any vector (x,y,z) can be written as a linear combination of
them.

© Copyright Univers idad Europea. Todos los derechos reservados

 Generators of vector spaces (or Spanning Sets)

Exercises:

Check if the set (1,2), (-1,1) and (2,0) spans the vector space ℝ𝟐

Find out if vectors (3,-1,2) and (1,0,-1) are a generating set of ℝ𝟑

© Copyright Univers idad Europea. Todos los derechos reservados

 Generators of vector spaces (or Spanning Sets)

A vector space can have many different generating systems.

There may be generating sets where some vectors are “not

necessary".
For example, if we have a set with four vectors in ℝ𝟑 , three of
them would really be enough to generate all the space.

There may be vector spaces that are too "large".

✓ A vector space V is of finite type if it is generated by a
finite number of vectors.
✓ That is, if there is a generator set 𝑺 = 𝒗𝟏 , 𝒗𝟐 , … , 𝒗𝒓 being r
a finite number.

© Copyright Univers idad Europea. Todos los derechos reservados

 Basis of a vector space

A basis of a vector space is:

✓ A set that generates that space,
✓ and no vector of the basis can be written as a linear
combination of the others (they are all linearly independent,
there are no “redundant” vectors).

If V is a finite vector space, and B is a vector set of V:

✓ B is a basis if and only if every vector of V can be expressed
in a unique way as a linear combination of the vectors of B.

© Copyright Univers idad Europea. Todos los derechos reservados

 Basis of a vector space

The set 1,0 , 0,1 is a basis of ℝ𝟐

The set 1,0,0 , 0,1,0 , 0,0,1 is a basis of ℝ𝟑

Those particular sets are the canonical bases of ℝ𝟐 and of ℝ𝟑

and the vectors are normally denoted as 𝒆𝒋

Exercises:

Study whether the set of vectors 1,1,1 , 0,1,1 , 0,0,1 is a basis of ℝ𝟑 .

What about the set of vectors 1,1,1 , 1,0,1 , 0,1,1 ?

© Copyright Univers idad Europea. Todos los derechos reservados

 Basis of a vector space

Every vector of a vector space V can be expressed as:

𝒏
𝒘 = 𝜶𝟏𝒗𝟏 + 𝜶𝟐𝒗𝟐 + ⋯ + 𝜶𝒏𝒗𝒏 = ෍ 𝜶𝒊 𝒗𝒊
𝒊=𝟏

where 𝐵 = 𝑣Ԧ1, 𝑣Ԧ2 , ⋯ , 𝑣Ԧ𝑛 is a basis of vector space V.

The values of 𝛼𝑖 are the coordinates of vector 𝑤 with

respect to B.

They are denoted by 𝑤𝐵 , or simply 𝑤 when the basis is canonical.

This representation is unique.

© Copyright Univers idad Europea. Todos los derechos reservados

 Unique Representation Theorem
Let V be a vector space and 𝑩 = 𝒗𝟏 , 𝒗𝟐 , ⋯ , 𝒗𝒏 a basis of that space.

For every vector 𝒘 that belongs to V there is exactly one

representation in terms of linear combination of the vectors in B.

𝒏
𝒘 = 𝜶𝟏𝒗𝟏 + 𝜶𝟐 𝒗𝟐 + ⋯ + 𝜶𝒏𝒗𝒏 = ෍ 𝜶𝒊 𝒗𝒊
𝒊=𝟏

Exercises:

Find the coordinates of the vector (3,-2,1) with respect to basis B=

1,1,0 , 1,0,1 , 0,1,1 .

© Copyright Univers idad Europea. Todos los derechos reservados

 Steinitz’s Theorem (or lemma)

From a basis 𝑩 = 𝒗𝟏, 𝒗𝟐 , ⋯ , 𝒗𝒏 of a vector space 𝑽, given a set L of

linearly independent vectors L = 𝒘𝟏, 𝒘𝟐, ⋯ , 𝒘𝒑 with 𝑝 < 𝑛, there
is at least one subset T of the basis, sized n−p such as:

The union of T with L 𝒘𝟏, 𝒘𝟐, ⋯ , 𝒘𝒑, 𝒗𝒑+𝟏 , ⋯ , 𝒗𝒏 is also a

basis of V.

Exercises:
Consider B= 𝒗𝟏 , 𝒗𝟐 , 𝒗𝟑 , basis of ℝ𝟑 , being 𝒗𝟏 = 1,0,0 , 𝒗𝟐 = 0,1,0 and
𝒗𝟑 = 0,0,1 , L = 𝒘𝟏, 𝒘𝟐 being 𝒘𝟏 = 1,1,0 and 𝒘𝟐= 1,1,1 and T =
𝒗1 = 1,0,0 . Prove that 𝑻 ‫ 𝑳 ڂ‬is also a basis of ℝ𝟑.
If T = 𝒗3 = 0,0,1 , is 𝑻 ‫ 𝑳 ڂ‬a basis of ℝ𝟑 ?

© Copyright Univers idad Europea. Todos los derechos reservados

 Corollaries to Steinitz’s Theorem: Dimension
1) If 𝒗𝟏 , 𝒗𝟐 , … 𝒗𝒏 is a basis of 𝑽 and if 𝒘𝟏 , 𝒘𝟐 , … , 𝒘𝒑 are linearly
independent vectors , then 𝒑 ≤ 𝒏.

2) If 𝑽 has a basis with 𝒏 vectors, then all bases of 𝑽 must have

𝒏 vectors.

Vector Space Dimension:

number of vectors that a basis has.

In a vector space of n dimensions:

3) Every set of 𝒏 + 𝟏 vectors of 𝑽 it is linearly dependent.

4) Every set of 𝒏 linearly independent vectors is a basis of 𝑽.

© Copyright Univers idad Europea. Todos los derechos reservados

Basis change (1)
A given vector space can have several different basis.

We can have a vector 𝒗 as linear combination of a given basis

𝑩 = 𝒗𝟏 , 𝒗𝟐, ⋯ , 𝒗𝒏 of space 𝑽 and as linear combination of a second
basis 𝑩′ = 𝒗′𝟏 , 𝒗′𝟐 , ⋯ , 𝒗′𝒏 of that same space.

Basis change matrix:

𝑀𝐵→𝐵′ = 𝑣′
Ԧ1 𝐵 𝑣′
Ԧ2 𝐵 𝑣′
Ԧ3 𝐵 … 𝑣′
Ԧ𝑛 𝐵

Where columns are the new basis B’ vectors coordinates in old basis B.

26
Basis change (2)
Example:

Given the basis 𝑩′ = 𝒃𝟏 , 𝒃𝟐 of ℝ𝟐 , formed by 𝒃𝟏 = (2, 1) and 𝒃𝟐 = (-1, 1),

and the vector 𝒗 = 4, 5 , expressed in the canonical basis, find the
coordinates of 𝒗 in 𝑩′.

Let’s apply definition of basis:

4 2 −1 2 −1 𝑎
𝒗= =𝑎 +b = . Solving this system, of equations, we
5 1 1 1 1 𝑏
3
obtain a = 3, b = 2 so the coordinates of 𝒗 in the new basis 𝑩′ are ,
2
2 −1
and the matrix of Basis change 𝑀𝐵→𝐵′ =
1 1

27
Vector coordinates change

Change of vector 𝒗 coordinates can be now obtained as a matrices

product:
𝑣Ԧ𝐵 = 𝑀𝐵→𝐵′ 𝑣Ԧ𝐵′
−1
𝑣Ԧ𝐵′ = 𝑀𝐵→𝐵′ 𝑣Ԧ𝐵

being 𝑣Ԧ𝐵 vector in basis B and 𝑣Ԧ𝐵′ in basis B’.

Exercise: Given vector 𝑥Ԧ = 3, 2 𝐶 and given basis 𝐵 = 𝑢1 , 𝑢2 , such as

𝑢1 = −𝑒Ԧ1 and 𝑢2 = 𝑒Ԧ2 , compute 𝑥Ԧ coordinates in canonical basis and in 𝐵
basis.

28
 Vector Subspace

It is a subset of a given vector space, where it applies that, for

any pair of vectors 𝑎,
Ԧ 𝑏 of that subset, and for any scalar j, k:
✓ 𝒂 + 𝒃 belongs to the subset
✓ k𝒂 belongs to the subset

✓So, in consequence:
✓ j𝒂 + 𝑘𝒃 belongs to the subset: any linear combination of the
vectors of the subset belongs to the subset
✓ Null vector belongs to the subset.

It is, likewise, a vector space (all axioms of vector spaces are

fulfilled).

© Copyright Univers idad Europea. Todos los derechos reservados

 Vector Subspace
✓ The following vector subspaces of a given space 𝑽 are called trivial
subspaces):
𝑽

𝟎
✓ Given a particular set of vectors of a vector space V that includes
𝟎 , it holds that all posible linear combinations of this set of
vectors form a subspace of V.

✓ Dimension of a vector subspace U: It is the range of the set of

vectors that generates (by means of a linear combination) the
subspace. It’s equal to the number of vectors belonging to any
basis of the subspace.
dim (U) ≤ dim(V)

© Copyright Univers idad Europea. Todos los derechos reservados

 Vector Subspace

Exercises:

Check if 𝑆 = 𝑎, 0 /𝑎 ∈ ℝ is a vector subspace of ℝ2 .

Check if 𝐿 = 𝑥, 𝑦, 𝑧 ∈ ℝ3/𝑥 − 2𝑦 + 3𝑧 = 0 is a vector subspace of

ℝ3 and find a basis of L.

Check if 𝑀 = 𝑥, 𝑦, 𝑧 ∈ ℝ3/𝑥 − 2𝑦 + 3𝑧 = 1 is a vector subspace of

ℝ3 .

© Copyright Univers idad Europea. Todos los derechos reservados

 Equations of a vector subspace

Implicit equations:

A generic vector is written in the form of equations showing the

relationship of coordinates:

𝑎𝑥 + 𝑏𝑦 + 𝑐𝑧 = 0

𝑎′ 𝑥 + 𝑏 ′ 𝑦 + 𝑐 ′ 𝑧 = 0

© Copyright Univers idad Europea. Todos los derechos reservados

 Equations of a vector subspace

Parametric equations:

A generic vector is written as equations as a function of

parameters (α,β)
𝑥 = α𝑥1 + 𝛽𝑥2
𝑦 = α𝑦1 + 𝛽𝑦2
𝑧 = α𝑧1 + 𝛽𝑧2

being 𝐵 = (𝑥1, 𝑦1, 𝑧1), (𝑥2, 𝑦2, 𝑧2) a subspace basis.

© Copyright Univers idad Europea. Todos los derechos reservados

 Equations of a vector subspace

Generator system/Basis:

A subspace U can be denoted either by a generator system

(using <> signs) o by a basis.

𝑆 = < 𝑣Ԧ1 , 𝑣Ԧ2 , … , 𝑣Ԧ𝑛 >

𝐵𝑈 = 𝑣Ԧ1 , 𝑣Ԧ2 , … , 𝑣Ԧ𝑛

34
 Equations of a vector subspace

Exercises:

Write the implicit and parametric equations of the subspace of ℝ3

generated by the basis of vectors: 𝑆 = −1,2,3 , 4, −5,1

Write the parametric and implicit equations of the subspace of ℝ4

generated by the basis of vectors: 𝑆 = −1,1,0,0 , 1,0,1,0 , −1,0,0,1

© Copyright Univers idad Europea. Todos los derechos reservados

 Intersection Subspace (1)
The set of all vectors belonging to two vector subspaces of V:

𝑼𝟏 ∩ 𝑼𝟐 = {𝒘 ∈ 𝑽/𝒘 ∈ 𝑼𝟏 , 𝒘 ∈ 𝑼𝟐 }
It is, likewise, a vector subspace.

Examples:
Calculate the implicit and parametrics equations of the
intersection of subspaces 𝑈1 and 𝑈2, 𝑈1 being spanned by 𝑈1 =
1,1,0 , 0,1,1 and 𝑈2 : 𝑥 + 𝑦 + 𝑧 = 0
We first calculate implicit equation of 𝑈1 , being 𝑈1 : x – y + z = 0
and we already know 𝑈2 : x + y + z = 0, so the implicit equations
of the intersection are both.
For the parametric equations, we transform this system of two
equations wiht three unknowns into x= - y=  z=  than
corresponds to a vector 𝒖 = (- ) or 𝒖 =  (- ) that spans a
subspace that is a line with basis: B = −1,0,1
© Copyright Univers idad Europea. Todos los derechos reservados
 Intersection Subspace (2)
Examples:

Calculate the intersection of subspaces spanned by:

𝑈1 = 1, −2,3 , −1,0,1 and 𝑈2 = 1,2,2 .

There are two ways of answering the example:

• 𝑈1 is a plane generated by 1, −2,3 , −1,0,1 and 𝑈2 is a line
generated by 1,2,2 . Their intersection can be either the zero
vector or the line itself (since all subspaces have to contain the
origin). So, if the three vectors are independent, the only
intersection can be the point (or vector): 0,0,0 . Calculating the
1 −1 1
rank of the matrix −2 0 2 , we have that r(A)=3, the vectors
3 1 2
are independent ant the only possible solution for the
homogeneus system is 0,0,0 .

© Copyright Univers idad Europea. Todos los derechos reservados

 Intersection Subspace (3)

• The other way is calculating the implicit equations for both

subspaces: for 𝑈1 , being 𝑈1: x + 2y + z = 0, and for 𝑈2 , they are:
2x - y = 0 and 2x - z = 0, and solvig the system formed by the
three equations we get that the only possible solution for the
homogeneus system is 0,0,0 .

© Copyright Univers idad Europea. Todos los derechos reservados

 Sum Subspace
The set of all vectors of V obtained by adding two vectors, each one
belonging to a subspace of V, is a vector subspace:

𝑼𝟏 + 𝑼𝟐 = {𝒘 ∈ 𝑽, 𝒖 ∈ 𝑼𝟏, 𝒗 ∈ 𝑼𝟐 /𝒘 = 𝒖 + 𝒗}

Example:
Calculate the Sum subspace of 𝑼𝟏 = (𝑥 = α, 𝑦 = α + 𝛽, 𝑧 = 𝛽) and 𝑼𝟐
= 𝑥+𝑦+𝑧 =0 .
Let’s find the out some basis for each subspace. For 𝑼𝟏 , a generic
vector can be written: 𝒖 = (α, α + 𝛽, 𝛽) = α 1,1,0 + 𝛽 0,1,1 so clearly
a basis is B1= (1,1,0), (0,1,1) . For 𝑼𝟐 we generate two independent
vectors, for example: 𝒗 =(-1,1,0) and 𝒘 =(-1,0,1),so B2=
(−1,1,0), (−1,0,1) so the Sum subspace will be spanned by these four
vectors. They cannot be independent, so let’s find out three of them
that are independent:

© Copyright Univers idad Europea. Todos los derechos reservados

 Sum Subspace

For example, 𝒖 = (1,1,0), (0,1,1) and (-1,1,0). Let’s check the

determinant:

1 0 −1
1 1 1 = - 1-1 ≠ 0 , so the three vectors are a basis of the Sum
0 1 0
subspace of ℝ𝟑 , so it corresponds to the entire ℝ𝟑 .

This example shows that the dimension of the Sum subspace is not
the sum of the dimensions of each subspace, instead:

𝒅𝒊𝒎 𝑼𝟏 + 𝑼𝟐 = 𝒅𝒊𝒎 𝑼𝟏 + 𝒅𝒊𝒎 𝑼𝟐 − 𝒅𝒊𝒎 𝑼𝟏 ∩ 𝑼𝟐

being 𝑼𝟏 and 𝑼𝟐 two subspaces of a vector space V. This leads us to

the definition of a special case of sum

© Copyright Univers idad Europea. Todos los derechos reservados

 Sum Subspace

Direct sum (𝑼𝟏 ⊕ 𝑼𝟐 ) :

Subspaces 𝑼𝟏 and 𝑼𝟐 have only the zero vector as their
intersection.

𝒅𝒊𝒎 𝑼𝟏 ⊕ 𝑼𝟐 = 𝒅𝒊𝒎 𝑼𝟏 + 𝒅𝒊𝒎 𝑼𝟐

When 𝑼𝟏 ⊕ 𝑼𝟐 span the entire vector space, they are called
supplementary (or complemented or complementary) subspaces.

Exercise:
Let’s be 𝑈1 = 1,1,0 and 𝑈2 = 1,0,1 , −1,1,0 . Calculate sum and
intersection of 𝑈1 and 𝑈2 .
Is their sum a direct sum?
If exists, find out the decomposition of the vector 2,3,1 as a direct sum
of vector in 𝑈1 and 𝑼𝟐
© Copyright Univers idad Europea. Todos los derechos reservados
 Thank you for
your attention!

© Copyright Univers idad Europea. Todos los derechos reservados 42