You are on page 1of 11

MATHS 2 | Amethyst Minds

3️⃣
IIT M BS Data Science
Week
L3.1 Introduction to Vector Spaces
Vector Space
Definition:

Let V over Rbe an arbitrary nonempty set of objects.

For which two operations are defined:

Addition (+): For each pair of elements v1 and v2 in V ,


​ ​

there's a unique element v1 ​


+ v2 .

Scalar Multiplication (.): For each scalar c(Scalar: Numbers


used for multiplication) and vector vin V , there's a unique
element c.v.

"Dot" (.) is often suppressed and written as cv.

Axioms: Certain conditions that all objects v1 , v2 , vin V and all


​ ​

scalars cmust satisfy.

Vector Space: If all axioms are met, V is termed a vector space.

Vectors: Objects in V are termed vectors.

MATHS 2 | Amethyst Minds 1


Vectors in Rn 

Here, Set Rn consists of n-tuples of real numbers.

Again for which two operations are defined:

Addition: coordinate-wise.

Scalar multiplication: coordinate-wise.

Some Vectors in Rn 

Matrices as Vector Spaces:

Set Mm×n represents all m × nmatrices, where each entry is a


real number.

Again two operations are defined:

Addition: (A + B)ij ​ = Aij + Bij .


​ ​

Scalar Multiplication: (cA)ij ​ = C(A)ij . ​

Linear Combinations: Any matrix in Mm×n can be expressed as a


linear combination of other matrices in Mm×n using matrix ​

addition and scalar multiplication.

Some Matrices in Vector Space:

MATHS 2 | Amethyst Minds 2


Axioms for Vector Space
Axioms of a Vector Space:
Let v1 , v2 , v′ be vectors in Rn and a, b
​ ​ ∈ R.
1. v1 + v2 = v2 + v1 (Commutative).
​ ​ ​ ​

2. (v1 + v2 ) + v3 = v1 + (v2 + v3 )(Associative).


​ ​ ​ ​ ​ ​

3. There exists a zero vector such that v + 0 = v.


4. For every v, there exists a v′ (negative of ) such that v + v′ = 0.
5. 1 × v = v.
6. Scalar multiplication properties:

a. a(bv) = (ab)v.
b. a(v1 + v2 ) = av1 + av2 (Distributive).
​ ​ ​ ​

c. (a + b)v = av + bv.
If a set V along with two operations satisfies these axioms above,
then is considered a vector space, and the elements within are
referred to as vectors.

Closed under an operation:

Set is closed if applying a specified operation to any two elements


produces an element in the set.

The operation can be addition, subtraction, multiplication, division,


or any other mathematical operation.

Ensures that the result of the operation does not take you outside
the set.

Homogeneous System of Linear Equations


A system of linear equations in the form Ax = 0.
'A' is a matrix of size m×n (m equations and n variables/unknowns).

Solution Set (V):

MATHS 2 | Amethyst Minds 3


V represents the set of all solution vectors (x)that satisfy Ax =
0.
This set V is a subset of Rn.

Key Properties:

Addition within V:

If v and w are two vectors from V (i.e., both are solutions to Ax


= 0), their sum (v + w) is also in V.

A(v + w) = A(v) + A(w) = 0 + 0 = 0.


Scalar Multiplication within V:

If v is a vector from V, and c is a scalar, then the scalar


multiplication c*v remains in V.

A(cv) = cA(v) = c × 0 = 0.


So addition and scalar multiplication on Rn restricts to the solution
set, Hence it is a vector space.

A subspace means it inherits properties of a vector space from its


parent space (here, Rn) and satisfies the required axioms.

V being a subspace is established by showing that it remains


closed under addition and scalar multiplication.
Non Examples
Not all definitions of operations (like addition or scalar multiplication)
necessarily lead to a structure that satisfies all axioms of a vector
space.

Commutativity Violation (Axiom i):

L3.2 Some properties of vector spaces


Cancellation Law and Its Corollaries

MATHS 2 | Amethyst Minds 4


Cancellation law states:

If xi ​
+ yi = zi + yi , then xi = zi .
​ ​ ​ ​ ​

Proved v1 + 0 = v2 + 0, leading to v1 = v2, used the fact that v3 + v3' =


0 (axiom).

Corollaries

Vector 0 is unique in a vector space.

The vector v' described in axiom 4 is unique and referred to as


minus v.

Proved uniqueness of vector 0 and vector v' using cancellation


law.
Properties of Vector Spaces
Three properties in any vector space V:

1. 0 times v = 0 for each v.

MATHS 2 | Amethyst Minds 5


2. c times v = -(cv) = c times (-v).

3. c times 0 = 0 for each scalar c in R.


Application Examples
Stock Taking

Used a grocery shop example with five items.

Considered the quantities in stock, buyer demands, and new stock


arrivals.

Treated this as an abstract vector space with addition and scalar


multiplication.

Highlighted the interpretation of negative and fractional quantities.

Affine Flats

Defined an affine flat as a plane parallel to the XY plane.

Discussed scalar multiplication by projecting onto the XY plane.

Explained addition by projecting points onto the XY plane and


using the parallelogram law.

Emphasized the geometric perspective in vector space operations.


L3.3 Linear Dependence
Linear combination
Given vectors v1 , v2 , vn and coefficients a1 , a2 , an , with there linear
​ ​ ​ ​ ​

combination you can form a new vector as ∑ ai vi . ​ ​

Example: 2 times (1, 2) + 1 times (2, 1) = (4, 5).

MATHS 2 | Amethyst Minds 6


Any vector in a set can be represented as a linear combination of the
other vectors in the set, including the ability to express the zero vector
using non-zero coefficients.

for example, 3(2,1) + 2(-2,3) - (2,9) = (0,0).


Linear Dependence
A set of vectors v1 , v2 , vn is linearly dependent if there exist scalars
​ ​ ​

a1 , a2 , an (not all ai are 0) such that ∑ ai vi = 0.


​ ​ ​ ​ ​ ​

Example 1: (1, 2) and (2, 1) are linearly dependent because

2(1, 2) - 2(2, 1) = (0, 0).

Example 2: (2, 3, 7) is linearly dependent on (2, 1, 2) and (3, 0, 1)


because

2(2, 1, 2) - 3(3, 0, 1) + 0.5(10, -4, -2) = (0, 0, 0).

Note: If a set is linearly dependent, any superset of that set is also


linearly dependent.

Linearly dependent vectors often form a line, plane, or higher-


dimensional subspace.

Scalar multiples of vectors can be linearly dependent.


L3.4 Linear Independence - Part 1
Linear Independence

MATHS 2 | Amethyst Minds 7


Recall that linear Dependence means a1 v1 ​ ​ + a2 v2 ...an vn = 0, with
​ ​ ​ ​

at least one ai non-zero.


A set of vectors v1 , v2 , vn is Linearly Independent if the only way a


​ ​ ​

linear combination equals 0 is when all coefficients ai are 0. ​

Some Examples
Two non-zero vectors in R2 

Zero vector in a set:

When two non-zero vectors are linearly independent, they cannot be


multiples of each other.

MATHS 2 | Amethyst Minds 8


Linear independence of three vectors:

If three vectors are linearly dependent, at least one can be


expressed as a linear combination of the others.

If three vectors are linearly independent, none can be expressed


as a linear combination of the others.

Example in R3:

The only solution to a linear combination equating to zero is when


all coefficients are zero.

MATHS 2 | Amethyst Minds 9


L3.5 Linear Independence - Part 2
Checking Linear Independence in Rm 
To determine if vectors v1 , v2 , ...vn in Rm are linearly independent,
​ ​ ​

we need to see if a linear combination of these vectors results in the


zero vector only when all coefficients are zero.

To check if vectors are linearly independent using a system of


equations, ensure that the homogeneous system V x = 0only has
the trivial solution (all zeros), where columns of V represent the
vectors.

Methods to determine linear independence


Scalar Multiples

If one vector is a scalar multiple of another, they're dependent.

In contrast, if two vectors are not scalar multiples of each other, it


doesn't automatically guarantee that they are linearly independent,
but it's a necessary condition for independence.

MATHS 2 | Amethyst Minds 10


Matrix Determinant

A set of vectors is linearly independent if and only if the


determinant of the matrix formed by these vectors as columns is
not zero.

For nvectors in Rn , calculate the determinant of the matrix they


form. A non-zero determinant indicates independence.

If the determinant of this matrix is 0, then the columns (vectors)


are linearly dependent.

Gaussian Elimination

Apply Gaussian elimination to the system formed by the vectors. A


row of zeros in the reduced form indicates dependence.

Dimension Comparison

If the number of vectors is greater than the dimension of the space


they are in, then the vectors are linearly dependent.

MATHS 2 | Amethyst Minds 11

You might also like