0% found this document useful (0 votes)
22 views26 pages

18bma61c U1

Uploaded by

Stalin Mandal
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
0% found this document useful (0 votes)
22 views26 pages

18bma61c U1

Uploaded by

Stalin Mandal
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/ 26

Unit I

Continuity and inverse images of open and closed sets.

Inverse image
Let f : S→T be a function from a set S to a set T.If Y is
a subset of T,the inverse image of Y under f, denoted by f-
1(Y) is defined to be the largest subset S which f maps into

Y.
f-1(Y)= {x: x  S, f(x)  Y}

Result:
Let A and B be subsets of T. then
A  B  f-1(A)  f-1(B)

Proof:
Let x  f-1(A)
 f(x)  A by defn of f-1(A)
 f(x)  B as A  B
 x  f-1(B)
 f-1(A)  f-1(B)
Theorem:
Let f : S→T be a function from S to T.
If X  S and Y  T then we have
(a) X = f -1(Y)  f(X)  Y
(b) Y= f(X)  X  f-1(Y)

(a)
Given X = f-1(Y)

Let y  f(X)
 y= f(x) for some x  X
 y= f(x) for some x  f-1(Y)
 y  Y by definition of f -1(Y)
 f(X)  Y
X=f-1(Y)  f(X)  Y
Given Y= f(X)
Let x  X
 f(x)  f(X)
 f(x)  Y since Y=f(X)
 x  f-1(Y)
Therefore X  f-1(Y)
Hence Y=f(X)  X  f-1(y)
Thus
f(f-1(Y))  Y and X  f-1(f(X))

Theorem:
Let f : S →T be a function from one metric space (S,ds)
to another metric space (T,dT) then f is continuous on S iff
for every open set Y in T, the inverse image f -1(Y) is open
is S.

Proof:
f is continuous  inverse image of every open set is
open under f
 let f be continuous on S.
Let Y be open in T.
Claim:
f-1(Y) is open in S
Let p f-1(Y)
To prove
p is an interior point of f -1(Y)
Let f(p)=y
As f(p)  Y and Y is open, f(p) is an interior point of Y.
There exists an  >0 such that BT(y,  )  Y …(1)
Since f is continuous at p,there is an  >0 such that
f(BS(p,  )  BT(y,  ) ..(2)
Now,
Bs(p,  )  f-1(f(Bs(p,  ))
 f-1(BT(y,  ) using (2)
 f-1(Y) using (1)
 p is an interior point of f-1(Y)
 f-1(Y) is open in (as p is arbitrary)


Let inverse image of every open set be open
To prove f is continuous
Let p S
Then f(p)  T
Let f(p) =y
For every  >0, the ball BT(y,  )  T
 By hypothesis
f-1(BT(y,  ) is open in S.
and p f-1 (BT(y,  ))
 there exist  >0 such that
Bs(p,  )  f-1(BT(y,  ))
 f(BS(P,  )  f(f-1(BT(y,  ))
 BT(y,  )
 f is continuous at p. (since p is arbitrary)
 f is continuous on S
Hence the proof.

Theorem:

Let f : S→T be a function from one metric space (S,ds) to


another metric space (T,dT). Then f is continuous on S iff
for every closed set Y in T, the inverse image f-1 (Y) is
closed in S.
Proof:
Note that, f-1(T-Y) = S - f-1(Y)
f is continuous  inverse image of every closed set is
closed.
 Let f be continuous
To prove inverse image of every closed set is closed.
Let Y be closed in T
Claim: f-1 (Y) be closed in S.

Y is closed in T  T-Y is open in T


 f-1(T-Y) is open in S
 S - f-1(Y) is open in S
 f-1(Y) is closed in S.
 inverse image of every closed set is closed.

 Let the inverse image of every closed set closed.


Claim: f is continuous
It is enough to prove that inverse image of every open
set is open.
Let Y be open in T
Then T-Y is closed in T
 f-1(T-Y) is closed in S (by hypothesis)
 S - f-1(Y) is open in S
 f is continuous
Remark:
The image of an open set under a continuous mapping
is not necessarily open.
Counter example:
Consider the function f : R→R
Define by f(x) = 1  x R

R is open in R.
Its image set f(R)={1}
{1} is not open in R because 1 is not an interior point of R.

(ii) The image of a closed set under a continuous mapping


need not be closed.
Consider the example
f : R→(-  /2,  /2) defined by f(x) = tan-1x
Then the image of a closed set is not closed in (-  /2,  /2)
Continuous functions on compact sets:

Definition of covering:-
A collection F of sets is said to be covering of a
given set S if
S   A F A
The collection F is said to cover S.
If F is a collection of open sets then F is called an open
covering of S.

Definition: Compact Set

A set S in Rn is said to be compact iff every open


covering of S contains a finite subcover.

Theorem:
Let f : S → T be a function from one metric space (S,ds)
to another metric space (T,dT). If f is continuous on a
compact subset X of S, then the image f(X) is a compact
subset of T. In particular, f(X) is closed and bounded in T.

Proof:
Given X  S is compact and f is continuous
Claim: f(X) is compact in T
Let F be an open covering of f(X)
(ie) f(X)   A F A
Since each A is open in T and f is continuous on S, each f -
1(A) is open in S. These sets f-1(A) form an open covering

of X.
For,
f(X)   A F A implies
f-1 (f(X))  f-1(  A F A) =  A F f-1 ( A)
This implies X   A F f-1 ( A) as X  f-1 (f(X))
Since X is compact, a finite number of open sets of {f -
1
(A)/A  F)} will cover X
(ie) X  f-1(A1)  f-1(A2)  …  f-1(AP)
 f(X)  f(f-1(A1)  f-1(A2)  …  f-1(AP))
= f(f-1(A1)  f(f-1(A2)  )……  f(f-1(AP))
 f(X) is closed and bounded, by theorem
“If S is subset of R , the following statements are
n

equivalent.
(1) S is compact.
(2) S is closed and bounded
(3) Every infinite subset of S has an accumulation
point in S”.

Hence the proof.

Bounded set :
A set X  Rn is a bounded set if there exists a  Rn
and r > 0, such that X  B(a,r).

Bounded function :
A function f : S → Rn is called bounded on S if
there is a positive number M such that
||f(x)||  M  x S.

Result:

Let f : S → Rn , then f is bounded on S iff f(S) is a


bounded set of Rn.
Proof:

Let f : S → Rn be bounded on S.
Then  M > 0 such that
||f(x)||  M  x S

Claim: f(S) is a bounded subset in Rn.


Since ||f(x)||  M ,  x S
f(S)  B(0,M) where B(0,M) is an open
ball with centre at origin and radius M.
i.e f(S) is a bounded subset in Rn

Let f(S) be a bounded subset of Rn
Then  a  Rn and r > 0 such that
f(S)  B(a,r)
Let x  S
||f(x)||= ||f(x)-0||
= ||f(x)-a+a-0||
 ||f(x)-a||+||a||
<(r+||a||)
= M1
This is true for every x  S.
 f is bounded on S.

Theorem:
Let f : S → Rn be a function from a metric space S to
the euclidean space Rn. If f is continuous on a compact
set X of S then f is bounded on X.

Proof:
Let f be a continuous function on the compact subset X
of S.
Then by theorem
“Let f : S → T be a function from one metric
space (S,ds) to another metric space (T,dT). If f is
continuous on a compact subset X of S, then the image
f(X) is a compact subset of T. In particular, f(X) is closed
and bounded in T”

f(X) is compact in Rn and f(X) is bounded in Rn


 f is bounded on X, by the previous result

Result:
If f is a real valued function bounded on X, then f(X) is
a bounded subset of R and so it has int f(X) and sup f(X)
and

Int f(X)  f(X)  sup f(X), x X


Theorem:
Let f : S → R be a real valued function from a metric
space S to the euclidean space R. Assume f is continuous
on a compact subset X of S. Then there exists a point
(p,q) in X such that f(p) = int f(X) ,and f(q) = sup f(X)

Proof:
Since f is a continuous function on the compact subset
S, f(X) is compact in R.
Also f(X) is closed and bounded in R
Since f(X) is bounded, m  f(x)  M with x X
where m=inf f(X) and M=sup f(X).
 every open ball with m as centre intersects f(X)
 m is an adherent point of f(x)
 m=f(p) for some p X
Similarly,
 every open ball with M a centre will also intersect
f(X)
 M is also an adherent point of f(X)
M  f(X) as f(X) is closed.
Let M=f(q) for some q X
Thus f(p)= inf f(X) and f(q) = sup f(X).

Theorem :Let f: S→T be a function from one metric


space(S, ds) to another metric space(T,dT). Assume that f
is one-to-one on S,so that the inverse function “f−1 “ exists.
If S is compact and if f is continuous on S, then f−1 is
continuous on f(S).

Proof :
Given f is a continuous function on the compact
space S. To prove, f−1: f(S)→S is continuous
We have, to prove that inverse image of every closed set
in S is closed in f(S), it is enough is prove that for every
closed set X in S, the image f(X )is closed in f(S).
Since X is closed and S is compact.

By theorem,

“Every closed subset of a compact space is compact.”


X is compact.
 f(X) is compact. By theorem,

“continuous image of a compact set is compact.”


 f(X) is closed by theorem,”
compact subset of a metric space is closed
and bounded”.

Topological Mappings (Homeomorphisms)

Let f : S →T be a function from one metric


space(S,ds) to another metric space (T,dT). Assume that f
is one-to-one on S, so that the inverse function f−1 exist. If
f is continuous on S and if f−1 is continuous on f(S) then f is
called a topological mapping or a homeomorphism.
In this case the metric spaces (S,ds) and (T,dT) are
said to be homeomorphic.

Note:
1. f is a homeomorphism then f-1 is also a
homeomorphism.
2. A homeomorphism maps open subsets of S onto
open subsets of f(S) and
3. It maps closed subsets of S onto closed subsets of
f(S)

Topological property:

Definition:
A property of a set that remains invariant under every
topological mapping is called a topological property.

Example:
The properties of being open, closed and
compact are topological properties.

Definition: Isometry

A function f : S→T which is one to one on S and which


preserves the metric is called an isometry.
(ie) If S is a isometry then,

ds(x,y) = dT (f(x),f(y)) for every x,y  S

If there is an isometry from (S,ds)→(f(S),dT) the two metric


spaces are called isometric.

Sign preserving property of continuous functions:

Theorem:
Let f be defined on an interval S in R. Assume that f is
continuous at a point c in S and that f(c)≠0. Then there
exist a one ball B(c,  ) such that f(x) has the same sign as
f(c) in B(c,  )  S.

Proof:
Since f is continuous at the point c there exist and  >0
for the given  >0 such that

f(c)-  < f(x) < f(c)+  x B(c,  )  S.


(1)
It’s given that c  S,

f(c)≠0 Suppose f(c) > 0

Then take  = f(c)/2 in (1)

 f(c) - f(c)/2 < f(x) < f(c) + f(c) /2

 1∕2 f(c) < f(x)< 3/2 f(c) x  B(c,  )  S

 f(x) > 0 for every

x  B(c,  )  S
Suppose f(c)
<0

 = -f(c)∕2 in
(1)

 f(c)+f(c)∕2 < f(c) < f(c) - f(c)∕2 ꓯ x B(c,  )  S

 3∕2f(c) < f(x) < f(c)∕2


Let f(c) = -m where m is positive
Then -3∕2 m <f(x)<-m∕2 ∀ x B(c,  )  S

 f(x) < 0 ∀ x B(c,  )  S


Hence the proof.

Bolzano’s theorem:

Let f be a real valued and continuous function on an


interval [a,b] in R and suppose that f(a)and f(b) have
opposite signs (i,e) f(a)f(b)<0 then there is atleast one
point c in open interval (a,b) such that f(c) =0
Proof:
Given that f(a) and f(b) have opposite signs.
Assume that f(a)>0 and f(b)<0

Let A = {x: x  [a,b], f(x)≥0}

Then A is non-empty. Since f(a)>0 , a  A


Also A is bounded
above by b. Therefore
A has a supremum.
Let c = sup A
Now a < c < b
To prove, f(c) =0
If f(c) ≠ 0 then by the sign preserving property of real
valued continuous function there is a 1-ball B(c,  ) in
which f has the same sign as f(c). If f(c) >0 there are
points x>c at which f(x) > 0, contradicting the definition of
C.

If f(c)<0, then c -  ∕2 is an upper bound for A again


contradicting the definition of c. Hence f(c) =0.

Theorem:
Assume f is real valued and continuous on a compact
interval S in R. Suppose there are two points α < β in S
such that f(α) ≠ f(β) then f takes every value between f(α)
and f(β) in the interval (α,β).
Proof:
Let α and β be such that f(α) ≠ f(β).

Let k be a number between f(α) and f(β)

Define: g: α,β]→R

g(x) = f(x)-k

then

g(α) = f(α)-k

g(β) = f(β)-k
then g(α) and g(β) have opposite sides

as f(α)<k<f(β) (or) f(β)<k<f(α)

By Bolzano’s theorem, there exists c  (α,β)such that


g(c) = 0
(i,e) f(c)-k = 0
 f(c) = k

 for every k in between f(α) and f(β) there exists a

c  (α,β) ϶ f(c) = k (k is arbitrary)


 f takes every value between f(α) and f(β).

Remark:
The continuous image of a compact interval S
under a real valued function is another compact interval
[inf f(S),supf(S)].
Proof:
By intermediate value theorem, the function f: S→R
defined on a compact interval takes every value between
f(a) and f(b) if S = [a, b]. This together with the theorem
“Let f: S→R be a real valued function from a metric space
S to Euclidean space R.
Assume that f is continuous on a compact subset of S.
Then there exists points p and q in x such that,f(p) = inf
f(X) and f(q) = sup f(X)” we have, f(S) = [ inf f(S), sup f(S)].

Connectedness
Definition:
A metric space S is called disconnected, if S = A∪B
where A and B are disjoint non-empty open sets in S. S is
connected if it is not disconnected.
Note:
A subset X of a metric space S is called connected if
when regarded as a metric subspace of S, it is a
connected metric space.

Examples:
1. Consider the metric space S = R-{0} with usual
Euclidean metric.

R-{0} = ( -∞,0)∪(0,∞)
Therefore R-{0} is disconnected.
2. Every open interval is connected.

Consider, T = {0,1} with the discrete metric


d(x, y) = {1 if x  y, 0 if x=y}
then T becomes the discrete metric space. The
possible subsets of T are  ,{0},{1},{0,1}. We
know

B(a, r) = {x: d(x, a)<r}

B (0,1∕2) = {0}, and B(1,1∕2) = {1}


Consider {0}. This is an open set of T.

Similarly,  ,{1},{0,1} are also open

Thus every subset of T is open.


 Every subset of T is closed.

3. The set Q of rational numbers regarded as a metric


subspace of Euclidean space is disconnected for Q
= A∪B where,
A consists of all rationals < √2 and B consists of all
rationals > √2. Also every ball in Q is
disconnected.
4. Every metric space S contains non-empty
connected subsets. In fact that for each p in S, the
set {p} is connected.

Definition:
A real valued function f which is continuous on a
metric space S is said to be two valued on f if f(S)  {0,1}
In other words a two valued function is a continuous
function whose only possible values are 0 and 1.
Note:
We usually consider the set T = {0, 1} with discrete
metric space T, where every subset is both open and
closed in T.

Theorem:
A metric space S is connected if and only if every two
valued function on S is constant.

Proof:
Assume S is connected . Let f be two valued on S.

Claim: f is constant
Let f−1({0}) = A and f−1({1}) = B be the inverse of the
subsets {0} and {1}. {0} and {1} are the open subsets of
the discrete metric space {0,1}. Since f is continuous,
both A and B are open in S.

Also AՈB = ∅

Hence S = A∪B where A and B are disjoint and open.

⟹ A = S and B = ∅ (or) B = S and A = ∅ as (S is not


disconnected)
⟹ f is constant on S.
conversely,

⟸ To prove, if every two valued function on S is


constant, then S is connected.

Suppose S is disconnected.

Then S = A∪B, where A and B are disjoint non-empty


open sets of S.
To prove there exists a two valued function on S, which is
not constant.
0 if x  A
Let f(x )={
1 if x B

Since A and B are non-empty,f takes both values 0 and 1.


So f is not a constant. f is continuous on S because the
inverse image of every open subset of {0,1} is open in S.
Thus f is two valued but not a constant.
⟹ ⇐ to the hypotheses
Hence S is connected
Hence the proof.

Continuous image of a connected set is connected


Theorem :
Let f: S→M be a function from a metric space S to
another metric space M. Let X be a connected subset of
S. If f is continuous on X, then f(X) is a connected subset
of M.

Proof :

Let g be a two valued function on f(X).


To prove : g is a constant.
Consider the composite function h: X→T defined
on X by h(x) = g(f(x)). Since, composition of two
continuous functions is continuous, h is
continuous. As h takes only the values 0 and 1 it
is two valued, As X is connected and h is two
valued on X, by previous theorem, h is a
constant.
 g is constant on f(X).

 f(X) is connected.

Hence the proof.

Result :
Every curve in Rn is connected :
Since an interval X⊆R is connected its continuous image
f(X) is connected. If f is real valued the image f(X) is
another interval. If f has values in Rn the image, f(X) is
called a curve in Rn and it is connected.

Theorem : Intermediate value theorem :


Let f be real valued and continuous on a connected
subset of Rn. If f takes two different values in S, say a
and b, then for each real c between a and b there exists a
point x in S, such that f(x) = c.

Proof :
f(S) being the continuous image of a connected set is
connected.
As f(S) ⊆ R and it is connected and it is an interval. Since
f takes the values a and b in S, f(S) is an interval
containing a and b.
 All values in between a and b are taken by f

 for a<c<b there exists x  S such that f(x) = c

Hence the proof.


Components of a metric space
Theorem :
Let F be a collection of connected subsets of a metric
space S such that the intersection T = ⋂𝐴∈𝐹 𝐴 is non-
empty. Then their union U = ⋃𝐴∈𝐹 𝐴 is connected.
Proof :
Given T = ⋂𝐴∈𝐹 𝐴 where F is a collection of connected
subsets of S. Since T is non-empty, let t∈T
t A for every A F →(1)
To prove, U is connected

Let f be a two valued function on U = ⋃A∈F A. It is


enough if we show that f is a constant, for S is connected
iff every two valued function on S is connected.

Let x  U be arbitrary.

 x  A for some A in F

 f(x) = f(t) ∀ x  A

Since A is connected and f is constant on A , by(1)∀ x U,

f(x) = f(t)
⇒f is constant on U.

Hence the proof.

Example :

Let A = (0, 1)
B = (1∕2, 2)
C = (-2, 1)

A B  C ≠ 

A∪B∪C = (-2, 2) which is connected.


Component of X :
Every point x in a metric space S belongs to atleast
one connected subset of S, namely {x}. The union of all
the connected subsets which contain x is connected and
is called the component of S and is denoted by U(x).

Note :
U(x) is the maximal connected subset of S which contains
x.
Example :

In the above example

Let A = (0, 1), B = (1∕2, 2), C = (-2, 1)

Each of A,B and C is connected consider the element 3∕4.

3∕4  A, 3∕4  B, and 3∕4  C

The largest connected set containing 3∕4 is A∪B∪C = (-2,


2) which is connected by above.

then,

U(¾) = (-2, 2)
Theorem :
Every point of a metric space S belongs to a uniquely
determined component of S. in other words the
components of S form a collection of disjoint sets whose
union is S.

Proof :
Since every point x  S belongs to atleast one
connected subset called {x}, we can say that x belongs to
atleast one component of S.
Union of components of S is S.
Components are disjoint.
Suppose x∈U1ՈU2 where U1 and U2 are two components of S.

 U1∪U2 is a maximal connected set containing X


condratracting the fact that U1 and U2 are maximal
connected sets containing X.

⇒U1ՈU2 = 

Hence the proof.

You might also like