Scribd
Upload
11 views8 pages

Lec 13

NPTEL Basic Linear Algebra

Uploaded by

neeru572010
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
11 views8 pages

Lec 13

NPTEL Basic Linear Algebra

Uploaded by

neeru572010
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

Basic Linear Algebra

Prof. Inder K. Rana


Department of Mathematics
Indian Institute of Technology – Bombay

Lecture - 13
Linear Span, Linear Independence and Basis - I

So let us begin our today’s lecture. We will just recall what we have done earlier. We have
looked at what is called the linear span of set of vectors. So given a subset 𝑆 of ℝ𝑛 , you define
what is called the linear span 𝐿(𝑆).

(Refer Slide Time: 00:48)

For a vector space 𝑉 in ℝ𝑛 , a set 𝐵 is a basis of 𝑉 if 𝐿(𝐵) = 𝑉 and there is no proper subset 𝐵0
of 𝐵 such that 𝐿(𝐵0 ) = 𝐿(𝐵) = 𝑉.

(Refer Slide Time: 02:15)


So basis of a vector space need not be unique. That is what we said, so given a vector space
there could be more than one set which form a basis. So the questions arise does every vector
space has a basis? How to find a basis of a given vector space and what is the relation between
any two basis of the vector space right? So given a vector space first question will like to answer
is does it have always have a basis right.

Second, does it how to find that basis given vector space and what is the relation between any
two basis of the vector space. So let us observe one thing to start with that basis if a vector
space if set 𝐵 is a basis, then 0 cannot belong to it right and this 0 cannot belong to the basis
right. If it is there anywhere because it is not going to do, so we will remove it. So basis will
always have because it is a minimal set.

So if 𝐵 is the set of vector which is a basis and contain 0, then we can delete 0 right because
any vectors we can take a linear combination 0 times 𝑣 + 0 times that will give you 0. So linear
span will always have 0, so if 0 is a part of a basis then it is not minimal, so we can always
remove it to make it a minimal right and definition of basis is the minimal set. So 0 will never
belong to the basis of a vector space okay.
(Refer Slide Time: 04:00)
So we want to show that every vector space has a basis. So the idea is following. So let us look
at the proof of this.
(Refer Slide Time: 04:13)

So theorem is if 𝑉 is obviously a nonzero vector space, then there exists a basis for 𝑉 right. So
let us look at a proof. See so 𝑉 is nonzero and 𝑉 is contained in ℝ^𝑛 right. So we start with
other set of generators right. So let 𝐵 contained in 𝑉 be a finite set such that a linear span of 𝐵
equal to 𝑉 right. We have got a set of generators and we want to check whether it is basis or
not. So what is the possibility?

So one possibility is case I that 𝐵 is minimal that is what is 𝐵 minimal means, there does not
exist any proper subset such that linear span of 𝐵0 is equal to 𝑉. Then, what will happen? That
means we are saying that this set itself is a basis. So that is 𝐵 is a basis, so what is case II? So
case I was it is minimal. So what is the second possibility? It is not minimal. So suppose 𝐵 is
not minimal, so let that means what?

If it is not minimal that means at least there is one vector right which you can remove. So let
𝑣 ∈ 𝐵 such that 𝐿(𝐵 ∖ 𝑣) = 𝐿(𝐵) = 𝑉.

So look at so it is not required that means what? So again two possibilities either this 𝐵 ∖ 𝑣 is
a basis, once we have removed that one vector which was not required so one possibility is
after removing one vector it becomes a basis. If not, again in so if not then there is there exists
some say 𝑢 belonging to 𝐵 ∖ 𝑣 which can be removed right like we removed earlier one vector.

So there is one possibility again that there is a vector which we can remove. So if you go on
doing this process what will happen right, the set 𝐵 is a finite set right, the basis is a finite set.
We removed one vector, we removed another vector right so eventually you cannot remove
everything, eventually there will be one vector left at least in the last stage if possibly then that
single vector will give everything right.
(Refer Slide Time: 08:41)

So after finite number of steps, so let us write after finite number of steps, we will get set 𝐵0
which is contained in 𝐵 such that 𝐿(𝐵0 ) = 𝑉 and 𝐵0 is minimal. That is 𝐵0 is a basis. So what
are we saying? We are saying that we have got a vector space which is generated by some set
of vectors 𝐵 that may not be minimal; I can make it minimal by removing one at a time which
are vectors which are not required.
(Refer Slide Time: 09:57)

(Refer Slide Time: 10:45)

(Refer Slide Time: 11:24)


So let us give some equivalent ways of describing what is the basis. So 𝑉 is a vector space
which is of course nonzero and 𝐵 is a set which is {𝑣1 , 𝑣2 , … , 𝑣𝑘 } contained in 𝑉. So we want
to give different ways of looking at when the set 𝐵 can be a basis. We have already defined
one definition as minimal set of generators. So the first one is 𝐵 is the basis that is it is the
minimal set of generators.

And we are going to show it is equivalent to saying that every element v of the vector space 𝑉
has a unique representation as a linear combination of elements of 𝐵. So let us prove that. Let
us prove that the I and II are equivalent.
(Refer Slide Time: 12:18)

(Refer Slide Time: 14:38)


(Refer Slide Time: 17:45)

(Refer Slide Time: 21:43)


This proves I is equivalent to II, I implies II and II implies I. So that was the statement we
wanted to show here.
(Refer Slide Time: 22:25)

So in the theorem, the first statement is 𝐵 is a basis, it is a minimal set of generators and were
shown it is equivalent to say that every element is representation as a linear combination of
elements of 𝐵 and that representation is unique, there is only one way of doing it.

You might also like