For a set X, if the distance map d is defined on X, we call (X,d) a metric space. We also say that X is metric.
Example.
Show that the followings are a metric on R.
d(x,y)=1+∣x−y∣∣x−y∣
d(x,y)=∣x−y∣pp≥1
Definition.
Let (X,d) be a metric space and A⊂X. We define the distance between the set and the point as
dist(x,A)=y∈Asupd(x,y).
Definition.
Let X be a vector space. A norm is a map ∥⋅∥:X→[0,∞) satisfying the followings.
∥x∥=0 if and only if x=0.
∥λx∥=∣λ∣∥x∥ for all x∈X and scalar λ.
∥x+y∥≤∥x∥+∥y∥ for all x,y∈X.
We call (X,∥⋅∥) A normed space.
Proposition.
A normed space is a metric space.
From now, we define a metric d(x,y)=∥x−y∥ from a norm. Indeed the converse of \cref{thm:2_norm2metric} is not true in general. Our question is that what additional condition is required so that a normed space has a metric.
Example.
Let (X,∥⋅∥) be a normed space and define d(x,y)=∥x−y∥.
Prove that d is a metric.
Show that
d(x+z,y+z)=d(x,y)
for all x,y∈X.
Show that
d(λx,λy)=∣λ∣d(x,y)
for all x,y∈X.
Proposition.
A metric space (X,d) is a normed space with its norm ∥x∥=d(x,0) if and only if
d(x+z,y+z)=d(x,y) for all x,y∈X.
d(λx,λy)=∣λ∣d(x,y) for all x,y∈X.
Definition.
An inner product is a map ⟨⋅,⋅⟩:X×X→R satisfying the followings.
⟨x,x⟩≥0 for all x∈X
⟨x,x⟩=0 if and only if x=0.
⟨λx,y⟩=λ⟨x,y⟩ for all x,y∈X and scalar λ.
⟨x,y+z⟩=⟨x,y⟩+⟨x,z⟩ for all x,y,z∈X
⟨x,y⟩=⟨y,x⟩ for all x,y∈X
We call (X,⟨⋅,⋅⟩) An inner product space.
Theorem.
{Cauchy-Schwarz Inequality}
For an inner product space (X,⟨⋅,⋅⟩),
∣⟨x,y⟩∣≤(⟨x,x⟩)21(⟨y,y⟩)21
Proof. Assume that x,y=0. For all α∈R, we have
0≤⟨αx+y,αx+y⟩=α2⟨x,x⟩+2α⟨x,y⟩+⟨y,y⟩
Therefore we derive that (⟨x,y⟩)2−⟨x,x⟩⟨y,y⟩≤0.
Proposition.
An inner product space is a normed space.
From now we define norm from inner product as ∥v∥=⟨v,v⟩. The thing is the triangle inequality could be derived from Cauchy-Schwarz inequality so that a norm is implied by an inner product.
Example.
Let (X,⟨⋅,⋅⟩) be an inner product space and define ∥x∥=⟨x,x⟩.
Prove that ∥⋅∥ is a norm.
Show that
2∥x∥2+2∥y∥2=∥x+y∥2+∥x−y∥2
for all x,y∈X.
Show that
⟨x,y⟩=41∥x+y∥2−41∥x−y∥2
for all x,y∈X.
Proposition.
A normed space (X,∥⋅∥) is an inner product space with its inner product
⟨x,y⟩=41∥x+y∥2−41∥x−y∥2
if and only if
2∥x∥2+2∥y∥2=∥x+y∥2+∥x−y∥2
for all x,y∈X. (Parallelogram Law)
Example.
Define
d(x,y)=∣x−y∣,
then (R,d) is a metric space. Without mentioning, we use this absolute difference as metric on R.
A is open if
for every x∈A, there exists r>0 such that B(x,r)⊂A.
A is closed if X∖A is open, that is, for every x∈X∖A, there exists r>0 such that B(x,r)⊂X∖A.
x∈A is an interior point of A if there exists r>0 such that B(x,r)⊂A.
An interior of A, denoted by A∘ is a set of all interior points of A.
B⊂A is open relative to A if there exists an open set U⊂X such that B=U∩A.
B⊂A is closed relative to A if there exists a closed set V⊂X such that B=V∩A.
The metric gives a nice topological properties to itself. A first step in constructing a topology is nothing but to define an open set. The open ball could be thought of as a circle in R2, a sphere in R3, or an open interval in R.
It is helpful to sketch an open ball and observe how it varies depending on the space and the metric.
Example.
Draw a unit open ball
B(0,1)={x∈R2∣dp(x,0)<1}
in R2 for p=1,2,3,∞ where
dp(x,y)=(∣x∣p+∣y∣p)p1.
Proposition.
Every open ball B(x,r) is open.
Proof. For all y∈B(x,r), take r1=r−d(x,y)>0, then
B(y,r1)⊂B(x,r)
since
for all z∈B(y,r1),
d(y,z)<r−d(x,y)⟹d(x,z)≤d(x,y)+d(y,z)<r.
Theorem.
Let U={Uα}α∈I be a collection of open sets, where I is an uncountable index set. Then the following sets are open.
α∈I⋃Uα
i=1⋂nUi
Proof. - For every x∈⋃α∈IUα, there exists Uβ∈U such that x∈Uβ.
Hence there is r>0 such that
$$ B(x, r) \subset U_\beta \subset \bigcup_{\alpha \in I} U_\alpha. $$ - For every $$ x \in \bigcap_{i=1}^n U_i $$, note that for $$ 1 \leq i \leq n $$ there exists $$ r_i>0 $$ such that $$ B(x, r_i) \subset U_i. $$ Take $$ r = \min \left \{ r_i \mid 1 \leq i \leq n\right \} $$, then
$$ x \in B(x, r) \subset \bigcap_{i=1}^n U_i. $$
Theorem.
Let V={Vα}α∈I be a collection of closed sets, where I is an uncountable index set. Then the following sets are closed.
α∈I⋂Vα
i=1⋃nVi
Proof. Trivial by De Morgan’s Law.
Example.
Let (X,d) be a metric space and A,B⊂X.
Show that A∘=A if and only if A is open.
Show that A∘=(A∘)∘.
Show that if A⊂B then A∘⊂B∘
Give an example such that A⊂B even though A∘⊂B∘.
Show that A∘∩B∘=(A∩B)∘.
Show that A∘∪B∘⊂(A∪B)∘.
Give an example such that A∘∪B∘=(A∪B)∘
% \subsection{Limit Points}
Definition.
Let (X,d) be a metric space and A⊂X.
x is a limit point of A whenever any open ball of x has a point in A other than x, that is,
B(x,r)∩(A∖{x})=∅
for all r>0.
We denote A′ a set of all limit points of A.
x is an isolated point of A if it is not a limit point of A.
A⊂X is closed if A contains all of its limit point, that is, A′⊂A.
A is perfect if all of the point in A is its limit point, that is, A=A′.
Proposition.
Let (X,d) be a metric space and A⊂X. Then X∖A is open if and only if A′⊂A.
Proof. \begin{steplist}
\item (⟹) Suppose not and let x∈X∖A be a limit point of A. Since X∖A is open,
there exists r>0 such that
B(x,r)⊂X∖A⟹B(x,r)∩(A∖{x})=∅,
which implies that x is an isolated point of A. (contradiction)
\item (⟸) Suppose not, then there exists x∈X∖A such that
B(x,r)⊂X∖A⟹B(x,r)∩(A∖{x})=∅
for all r>0, which implies that x is a limit point of A. (contradiction)
\end{steplist}
Proposition.
X and ∅ are both open and closed whenever X is a metric space.
Our next question is that whether there exists a metric space so that all of its subsets are both open and closed. The following is one example of such a space.
Example.
Let (X,d) be a metric space with discrete metric
d(x,y)={10(x=y)(x=y).
Prove the followings.
A⊂X is open.
A⊂X is closed.
Example.
Let (X,d) be a metric space and A,B⊂X.
Show that if A⊂B then A′⊂B′.
Give an example such that A′⊂B′ but A⊂B.
% \subsection{Closure and Boundary}
Definition.
Let (X,d) be a metric space and A⊂X.
The closure of A denoted by A is defined as
A=A∪A′.
The boundary of A, denoted by ∂A is defined as ∂A=A∩X∖A.
A is dense in X if A=X.
A is nowhere dense in X if (A)∘=∅.
A collection {Uα}α∈I of open subsets of X is A base for X if for all x∈X and every open set V containing x, there exists Uα such that x∈Uα⊂V.
X is separable if there exists a countable dense subset of X.
Proposition.
A is the smallest closed set containing A.
Proof. Consider a closed set V such that A⊂V⊂A. Since A′⊂V, V=A.
Corollary.
A=A if and only if A is closed.
Theorem.
Let (X,d) be a metric space and A⊂X. x∈∂A if and only if
B(x,r)∩A=∅andB(x,r)∩(X∖A)=∅
for all r>0.
Proof. \begin{steplist}
\item (⟹) Without loss of generality, it suffices to assume that x∈A. We observe
x∈A⟹B(x,r)∩A=∅
and
x∈X∖A⟹x∈(X∖A)′⟹B(x,r)∩(X∖A)=∅.
\item (⟸) For all r>0,
B(x,r)∩A=∅⟹x∈A
and
B(x,r)∩(X∖A)=∅⟹x∈X∖A
therefore x∈∂A.
\end{steplist}
Example.
Let (X,d) be a metric space. Prove the followings.
(A∘)c=Ac
(Ac)∘=(A)c
∂A=A∖A∘
As a result of previous example, we observe that A∘, ∂A, and (Ac)∘ are a partition of X, that is, these three sets are disjoint and A∘∪∂A∪(Ac)∘=X.
Example.
Let (X,d) be a metric space. Prove the followings.
∂A⊂Ac if and only if A is open.
∂A⊂A if and only if A is closed.
∂A=∅ if and only if A is both open and closed.
% \subsection{Bounded Sets}
Definition.
Let (X,d) be a metric space.
The diameter of A denoted by diamA, is defined as
diamA={sup{d(x,y)∣x,y∈A}0(A=∅)(A=∅).
A is bounded if and only if diamA<∞.
Example.
Let (X,d) be a metric space and A,B⊂X.
Show that if A is bounded then so be A.
Give an example such that A is not bounded even though A∘ is.
Sequences and Convergence
% \subsection{Convergence}
Definition.
Let (X,d) be a metric space. For a sequence (xn) in X, we say xnconverges to x∈X, denoted by
n→∞limxn=x
or
xn→x
if for every ε>0, there exists n0 such that
n≥n0⟹d(xn,x)<ε.
Theorem.
Let (X,d) be a metric space and A⊂X. A is closed if and only if
for every sequence (xn) in A that converges in X, limn→∞xn∈A.
Proof. (⟹) Suppose limn→∞xn=x∈/A, then
for all ε>0, there exists n0 such that
n≥n0⟹d(xn,x)<ε⟹B(x,ε)∩(A∖{x})=∅
therefore x∈A′. (contradiction)
(⟸) It suffices to show that A′⊂A. Suppose x∈A′, then
for all r>0,
B(x,r)∩(A∖{x})=∅
therefore we could construct a sequence xn∈X such that
for a given ε>0, there exists n0=min{n∈Nn1<ε} so that
n≥n0⟹xn∈B(x,n1)∩(A∖{x})
therefore limn→∞xn=x∈A.
Theorem.
For a sequence (xn) in a metric space (X,d), if
n→∞limxn=xandn→∞limxn=y
then x=y.
Proof. Let ε>0 be given. Then there exists n0 such that
n≥n0⟹d(xn,x)<2εandd(xn,y)<2ε,
which implies
d(x,y)≤d(x,xn)+d(xn,y)<ε
hence x=y.
Theorem.
Let (X,d) be a metric space and A⊂X. x∈A if and only if
there exists a sequence (xn) in A that converges to x.
Proof. It suffices to consider the case of x∈A∖A.
(⟹) Suppose x∈A∖A⊂A′, then
for all r>0, B(x,r)∩(A∖{x})=∅
therefore we could construct a sequence xn∈X so that
for all ε>0, there exists n0 such that
n≥n0⟹xn∈B(x,ε)∩(A∖{x}).
(⟸) Suppose that limn→∞xn=x, then
for all ε>0, there exists n0 such that
n≥n0⟹xn∈B(x,ε)
which implies that
B(x,r)∩A=∅
for all r>0.
Definition.
Let (X,d) be a metric space and (xn) be a sequence in X.
xn is bounded if {xn∣n∈N} is bounded.
For a strictly increasing sequence (nk) in N, we say (xnk) A subsequence of xn.
Theorem.
If x is a limit point of {xn∣n∈N}, then there exists a subsequence of xn which converges to x.
Proof. Since x∈{xn∣n∈N}′,
for all r>0,
B(x,r)∩({xn∣n∈N}∖{x})=∅
hence we could construct a subsequence xnk such that
for a given ε>0, take k0=min{k∈Nk1<ε} so that
k≥k0⟹xnk∈B(x,k1)∩({xn∣n∈N}∖{x})
therefore limk→∞xnk=x.
% \subsection{Cauchy Sequences}
Completeness
Definition.
Let (X,d) be a metric space and (xn) be a sequence in X.
xn∈X is a Cauchy sequence if
for all ε>0, there exists n0 such that
m,n≥n0⟹d(xm,xn)<ε.
X is complete if every Cauchy sequence in X converges to a point in X.
Example.
Show that (0,1] is not complete.
Proposition.
Let (X,d) be a metric space and A⊂X. If A is complete, then A is closed.
The converse of \cref{prop:2_complete_closed} is indeed not true, as the following example.
Example.
Show that Q is complete but not closed.
Theorem.
Every convergent sequence in a metric space is a Cauchy sequence.
Proof. Suppose (xn) in a metric space X be a sequence such that limn→∞xn=x∈X, then
for a given ε>0, there exists n0 such that
n≥n0⟹d(x,xn)<2ε.
By the triangle equality,
m,n≥n0⟹d(xn,xm)≤d(xn,x)+d(xm,x)<ε
therefore (xn) is Cauchy.
Theorem.
Every Cauchy sequence in a metric space is bounded.
Proof. Suppose (xn) be a Cauchy sequence in a metric space X. Then there exists n0 such that
m,n≥n0⟹d(xn,xm)<1.
Take M=sup{d(xm,xn)∣m,n≤n0}+1 then diam{xn∣x∈N}≤M.
Compactness
% \subsection{Compact Sets}
Definition.
Let (X,d) be a metric space.
A⊂X is sequentially compact if every sequence in A has a subsequence converges to a point in A.
A⊂X is limit point compact if every infinite subset B⊂A has a limit point in A.
A cover of A⊂X is a collection of sets {Uα}α∈I satisfying
A⊂α∈I⋃Uα.
A cover is an open cover if each element of it is open.
A⊂X is compact if every open cover of A has a finite subcover.
Example.
Show that (0,1)⊂R is not compact.
Proposition.
Let (X,d) be a metric space.
A⊂X is sequentially compact if and only if it is limit point compact.
Proposition.
Let (X,d) be a metric space and K⊂X be a compact set. Then K is closed.
Example.
Prove \cref{prop:2_seq_lim_cpt} and \cref{prop:2_cpt_cl}.
Proposition.
Let (X,d) be a metric space and K⊂X be a compact set. A⊂K is compact if it is closed.
Proof. Let {Uα}α∈I be an open cover of A. Then we note that {Uα}α∈I∪(K∖A) is indeed an open cover of K and there is an finite subcover since K is compact so we are done.
Example.
Indeed, if K is compact and A is closed then K∩A is compact.
Definition.
Let (X,d) be a metric space, A⊂X and {Uα}α∈I be an open cover of A.
r>0 is a Lebesgue number of {Uα}α∈I if for all B⊂A with diamB<r, there exists β∈I so that
B⊂Uβ.
Theorem.
Let (X,d) be a metric space and A⊂X. If A is sequentially compact, then there exists the Lebesgue number of each open cover of A.
Proof. Let {Uα}α∈I be an open cover of A and r>0 be its Lebesgue number.
\begin{steplist}
\item Suppose not, then there exists a sequence (xn)⊂A such that
B(xn,n1)⊂Uα
for all α∈I.
\item Since A is sequentially compact, there exists a subsequence xnk∈A of xn such that
k→∞limxnk=x∈A.
Let {Uα}α∈I be an open cover of A then we have for r1>0,
x∈B(x,r1)⊂Uβ
for some β∈I.
Further we have k0 such that
k≥k0⟹xnk∈B(x,2r1)andnk1<2r1.
\item Consequently we obtain
B(xnk,nk1)⊂B(xnk,2r1)⊂B(x,r1)⊂Uβ.
(contradiction)
\end{steplist}
Definition.
Let (X,d) be a metric space.
A⊂X is totally bounded if for all r>0, there exists finite set {x1,x2,⋯,xn}⊂X such that
A⊂i=1⋃nB(xi,r).
Proposition.
Let (X,d) be a metric space.
If A⊂X is totally bounded then it is bounded.
Example.
On the metric space (R,d) with the discrete metric
d(x,y)={10(x=y)(x=y),
Show that Z is bounded but not totally bounded.
Theorem.
Let (X,d) be a metric space.
If A⊂X is sequentially compact, then it is totally bounded.
Proof. Let A be sequentially compact and suppose not.
\begin{steplist}
\item Let r>0 be given. Then we can construct a sequence (xn)⊂A by taking some x1∈A and choosing
xn+1∈A∖k=1⋃nB(xk,r)
for all n≥1. We note that A∖⋃k=1nB(xk,r) is nonempty since A is not totally bounded.
\item Since A is sequentially compact, there exists a convergent subsequence (xnk) of (xn). Hence, there exists n0 so that
n≥n0⟹d(xn,x)<2r⟹xn∈B(x,2r)
but it is indeed unable by our construction. (contradiction)
\end{steplist}
Theorem.
Let (X,d) be a metric space.
A⊂X is sequentially compact if and only if it is complete and totally bounded.
Proof. This proof is left as an exercise.
% \subsection{Compactness in Metric Spaces}
The following theorem is one of the most important theorem in this course.
Theorem.
Let (X,d) be a metric space.
A⊂X is compact if and only if it is sequentially compact.
To prove the theorem, we first introduce the following lemma.
Lemma.
Let (xn)⊂A be a sequence such that every subsequence of it diverges.
For all n, there exists rn>0 such that
B(xn,rn)∩{xm∣xm=xn}=∅.
Proof. If not, then there exists a subsequence (xnk)⊂A such that
d(xnk,xk0)<k1
for all k and some k0, implying xnk→xk0. (contradiction)
Proof. [Proof of the theorem]
(⟹)
\begin{steplist}
\item Suppose not. Then there exists a sequence (xn)⊂A such that every subsequence of it does not converge. Without the loss of generality, we assume xi=xj if i=j.
\item \cref{lem:2_BW1} yields that there is no limit point of {xn∣n∈N}, so we observe {xn∣n∈N} is closed. Note that {xn∣n∈N} is compact since it is closed subset of the compact set A.
However we readily check
{B(xn,rn)}n∈N
is an open cover of {xn∣n∈N} with no finite subcover. (contradiction)
\end{steplist}
Proof.(⟸)
\begin{steplist}
\item Let {Uα}α∈I be an open cover of A. By \cref{thm:2_S_T} and \cref{thm:2_Lebesgue}, we observe A is totally bounded and there exists a Lebesgue number r>0 of {Uα}α∈I. Hence there exists a finite set
{y1,y2,⋯,ym}
so that
A⊂k=1⋂mB(ym,r).
\item Moreover for each B(yk,r) there exists α(yk)∈I so that
B(yk,r)⊂Uα(yk)
whence
{Uα(y1),Uα(y2),⋯,Uα(m)}
is an open cover of A.
\end{steplist}
% \subsection{Compactness in Finite-Dimensional Vector Spaces}
Theorem (Heine-Borel).
Let (X,∥⋅∥) be a finite-dimensional complete normed space.
A⊂X is compact if and only if A is closed and bounded.
Proof. (⟹) Obvious.
(⟸)
\begin{steplist}
\item By \cref{thm:2_B-W}, it suffices to show that A is sequentially compact. Let dimX=d and {e1,⋯,ed} be a basis for X. Let (xn)⊂A be a sequence then we have a representation
xn=ξ1(n)e1+⋯+ξd(n)ed.
\item Note that ∥xn∥ is bounded and we have the inequality
ci=1∑d∣ξi(n)∣≤i=1∑dξi(n)ei(n)=∥xn∥<∞
for some c>0.
Hence for a fixed i, the sequence (ξi(n)) is bounded so there exists a convergent subsequence (ξi(nk))→ξi.
\item Consequently we conclude there exists a subsequence (xnk) of (xn) so that
xnk→i=1∑dξiei
that is, A is sequentially compact.
\end{steplist}
\newpage
Summarising what we have leant, it is helpful to refer the following diagram. Note that if the space is not finite-dimensional complete normed, closed and bounded does not imply the compactness.
Let (X,d) be a metric space. A collection {Vα}α∈I of the sets in X has a finite intersection property if the any finite intersection of its element is nonempty.
Theorem.
Let (X,d) be a metric space. A⊂X is compact if and only if for any collection {Vα}α∈I with finite intersection property, the intersection
A∩α∈I⋂Vα
is nonempty.
Proof. (⟹) Suppose not then A∩Vβ=∅ for some β∈I. Let Uα=X∖Vα then indeed the collection {Uα}α∈I is the open cover of A. Since A is compact, there exists its finite subcover and this violates the finite intersection property of {Vα}α∈I.
(⟸) Indeed, the similar argument works so we leave it for an exercise for the readers.
Corollary.
Let (X,d) be a metric space and (Kn)⊂X be a decreasing sequence of nonempty compact sets in X, then
⋂n=1∞Kn
is nonempty.
Proof. Obvious.
% \subsection{Connected Sets}
Connectedness
Definition.
Let (X,d) be a metric space.
A⊂X is disconnected if there exists disjoint open sets U and V such that separateA, that is,