Acknowledgement

First, I extend my sincere gratitude towards my guide Dr. Pearl Sanchayeta Gogoi,
Assistant Professor, Mathematics Department,Arya Vidyapeeth College, for her continuous
effort and support, valuable suggestions and for being an infinite source of information guid-
ing well throughout my project.

Next, I would like to thank all the respected faculty members of Mathematics Department,
Arya Vidyapeeth College, for their shear guidance and motivation.

Lastly I offer my thanks to all my friends for being great help and encouragement throughout.

Date LEENA SARMA

GU ROLL NO. :PS-201-001-0029
GU REGISTRATION NO.: 17000615 of 2020-21
ARYA VIDYAPEETH COLLEGE, GUWAHATI-16

1
Contents

1 Chapter
Introduction And Preliminaries 5

1.1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

1.2 PRELIMINARIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

1.2.1 Metric Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

1.2.2 Vector Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

1.2.3 Norm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

1.2.4 Normed Linear Space . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

1.2.5 Metric Induced by Norm . . . . . . . . . . . . . . . . . . . . . . . . . 8

1.2.6 Banach Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

1.2.7 Inner Product space . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

1.2.8 Hilbert Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

1.2.9 Holder inequality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

1.2.10 Minkowski inequality . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

1.2.11 Cauchy Schwarz inequality . . . . . . . . . . . . . . . . . . . . . . . . 12

2 Chapter
Linear operators 13

2.1 Operator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

2.2 Linear operator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

2.2.1 Kernel of T . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

2.2.2 Rank of a linear operator . . . . . . . . . . . . . . . . . . . . . . . . . 14

2.3 Bounded Linear Operator . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

2.4 Continuous Linear Operator . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

2.5 Linear Functionals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

2
 2.5.1 Bounded linear functional . . . . . . . . . . . . . . . . . . . . . . . . 17

2.6 Normed spaces of Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

2.7 Dual Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

2.8 Isomorphism of norm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

2.9 Some Important Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

2.9.1 Hahn-Banach Theorem(Extension of linear functionals) . . . . . . . . 19

2.9.2 Hahn-Banach Theorem(Generalized) . . . . . . . . . . . . . . . . . . 19

2.9.3 Uniform Boundedness Theorem . . . . . . . . . . . . . . . . . . . . . 20

2.9.4 Open Mapping Theorem . . . . . . . . . . . . . . . . . . . . . . . . . 20

2.9.5 Closed Graph Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . 20

3 Chapter
Strong And Weak Convergence 22

3.1 Strong Convergence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

3.2 Weak Convergence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

3.3 Convergence in Hilbert space . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

3.4 Convergence of Sequence of Operators . . . . . . . . . . . . . . . . . . . . . 25

4 Chapter
Spectrum 26

4.1 Spectrum in finite dimensional normed space . . . . . . . . . . . . . . . . . . 26

4.1.1 Resolvent set: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

4.2 Spectrum in infinite dimensional normed space . . . . . . . . . . . . . . . . . 27

4.2.1 Resolvent Set: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

4.2.2 The Point Spectrum: . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

4.2.3 The Continuous Spectrum . . . . . . . . . . . . . . . . . . . . . . . . 28

4.2.4 The Residual Spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . 28

3
 4.2.5 Relation to eigenvalues . . . . . . . . . . . . . . . . . . . . . . . . . . 29

4.2.6 Spectrum of a Bounded Operator . . . . . . . . . . . . . . . . . . . . 29

4.2.7 Spectrum of an unbounded operator . . . . . . . . . . . . . . . . . . 30

5 Conclusion 31

6 Bibliography 32

4
 1 Chapter
Introduction And Preliminaries

1.1 INTRODUCTION

Emerging from classical analysis, functional analysis is that branch of mathematics which
generalises and provides a basis for a unified treatment bringing out essence of a problem by
clearing out unnecessary details, from various branches of analysis.It embodies the abstract
approach to analysis.The core of the branch is rooted in the study of spaces of functions and
the formulation of properties of transformation of functions.In a nutshell functional analysis
is the study of normed vector spaces and bounded operators, extending and highlighting the
interconnection between algebraic structures and distance structures.This area of mathemat-
ics has both intrinsic beauty and vast applications in many fields of mathematical sciences
including applied sciences as well.

Initially in this project I have tried as a beginner to introduce the basic concepts of func-
tional analysis followed by discussion of few topics relating to convergence i.e. strong and
weak convergence adjoining their important results and derivations as well.Furthermore, I
have included a short discussion on the topic spectrum of operators.

5
1.2 PRELIMINARIES

In calculus we study functions on a real line R.Though limiting process we see that on R we
have available a distance function say d,which refers a distance d(x, y) = |x − y| with each
pair of points.
In functional analysis we generalise R by an abstract set X and introduce a distance function
as well. Here we start by defining the very basics of functional analysis.

1.2.1 Metric Space

Let X be a non empty set. A metric space on X is a real valued function d : X × X → R

which satisfies the following conditions-

a) d(x, y) ⩾ 0 ∀ x, y ∈ X i.e. d is real valued,finite and non-negative.

b) d(x, y) = 0 if and only if x = y
c) d(x, y) = d(y, x) (symmetry)
d) d(x, y) ⩽ d(x, z) + d(z, y) (triangle inequality)
A metric space is a non empty set X equipped with a metric d on X, denoted by the pair
(X, d).
Examples- Real line R Set of all real numbers, taken with usual metric defined by:

d(x, y) = |x − y|

Euclidean Space Rn - Set of all ordered n tuples of real numbers i.e. x = (x1 , x2 , . . . , xn ), y =
(y1 , y2 , . . . , yn ) and the Euclidean metric is defined by:
p
d(x, y) = (x1 − y1 )2 + (x2 − y2 )2 + · · · + (xn − yn )2

1.2.2 Vector Space

Roughly speaking it is a set of elements called vectors. Two vectors can be added resulting
in a new vector and a vector can be multiplied with a scalar, so as to give a new vector. The
precise definition is as follows:
Let X be a non-empty set and F be a field. The set X together with two functions
+ : X × X → X called vector addition and ∗ : F × X → X called scalar multiplication

6
is called a vector space over F if it satisfies the following conditions:

V1) ∀x1 , x2 , x3 ∈ X, x1 + (x2 + x3 ) = (x1 + x2 ) + x3

V2) ∃ an element 0, such that ∀x ∈ X
x+0=0+x=x
V3) ∀x ∈ X, ∃ an element (−x) such that
(−x) + x = x + (−x) = 0
V4) x1 + x2 = x2 + x1 ∀x1 , x2 ∈ X
V5) 1x = x∀x ∈ X
V6) α(βx) = (αβ)x, ∀α, β ∈ F and x ∈ X
V7) (α + β)x = αx + βx, ∀α, β ∈ F and x ∈ X
V8) α(x1 + x2 ) = αx1 + αx2 , α ∈ F and x1 , x2 ∈ X
Examples-

i)Euclidean space Rn : Rn is a vector space over the field R

ii)Complex space Cn over the field C
iii)Space C[a, b]: Each point is continuous real valued function on [a, b]. The set of all these
function forms a real vector space with the algebraic operations
(x + y)(t) = x(t) + y(t)
(αx)(t) = αx(t)

1.2.3 Norm

A norm is a real-valued function || • || : X → R from a real or complex vector space to real

numbers satisfying the following properties:

N1) ||x|| ⩾ 0
N2) ||x|| ⇔ x = 0
N3) ||αx|| = |α|||x||
N4)||x + y|| ⩽ ||x|| + ||y||
where x and y are arbitrary vectors in X and /alpha is any scalar of the field F.

1.2.4 Normed Linear Space

Let X be a vector space over the field F. We know that norm is a function || • || : X → R
satisfying the above mentioned properties. A normed linear space X is a vector space with
a norm defined on it. A normed space is denoted by (X, || • ||
Examples:

7
i) R over the field R, if we define || • || : R → [0, ∞) by- ||x|| = x, x ∈ R

ii)Rn over R defined by- ||x|| = ( n(i=1) (xi )2 )(1/2)

P
p
= (x1 )2 + (x2 )2 + · · · + (xn )2

1.2.5 Metric Induced by Norm

The properties defining a norm:

N1) ||x|| ⩾ 0
N2) ||x|| = 0 ⇔ x = 0
N3) ||αx|| = |α|||x||
N4) ||x + y|| ⩽ ||x|| + ||y||
where x, y are arbitrary vectors in X and α is a scalar.
The first two defining properties of norm above are suggested and motivated by the length
of a vector x in elementary vector algebra,so it can be written as ||x|| = x. Furthermore,the
properties N1 and N2 states that all vectors have positive lengths except the zero vector
which has length zero. N3 displays that when a vector is multiplied by a scalar, its length
is multiplied by the absolute value of the scalar. N4 suggests the triangle inequality i.e. the
length of one side of a triangle cannot exceed the sum of the lengths of the other two sides.

Thus from the above mentioned reasons it can be concluded that:

• A norm on X defines a metric d on X which is given by: d(x, y) = ||x − y|| and is called
the metric induced by the norm.
• Every normed linear space is a metric space but every metric space is not a normed space.
Example: A discrete metric space is not a normed space.
(The above result can be easily proved.)

1.2.6 Banach Space

Banach space play a central role in functional analysis.

To know banach space we must be aware of few terms and definitions:

Convergence of a sequence in metric space: A sequence xn in a metric space X = (X, d)

is said to converge if there is an x ∈ X such that

8
 lim d(xn , x) = 0
n→∞

where x is called the limit of xn and can be written as:

lim xn = x
n→∞

orxn → x

Cauchy’s sequence in metric space: A sequence xn in (X, d) is said to be a Cauchy

sequence if for every ϵ > 0 there is a Nϵ such that

d(xm , xn ) < ϵ
for every m,n> Nϵ
Also it is known to us that every convergent sequence in a metric space is a Cauchy sequence.

Complete metric space: A metric space (X, d) is said to be complete if every Cauchy’s
sequence in (X, d) is convergent to some point in X.

Knowing the above facts, now Banach space can be defined.

Definition: Let X be a normed linear space and d be the induced metric space i.e. d(x, y) =
||x − y||. If X is complete i.e. complete in the sense that a Cauchy sequence of vectors always
converges to a well defined limit that is within the space, with respect to the metric d,then
X is said to be a complete normed linear space and a complete normed space is called a
Banach space.

Examples: 1) Euclidean space Rn is a banach space with the norm defined by-
 !(1/2) 
Xn
p
||x|| =  |ξi |2  = |ξ1 |2 + · · · + |ξn |2
i=1

i.e. Rn is complete and yields the metric

p
d(x, y) = ||x − y|| = (ξ1 − η1 )2 + · · · + (ξn − ηn )2

2) lp space:(p⩾ 1 be a fixed real number)

∞
X
lp = (x = (x1 , x2 , . . . ) : |xi |p < ∞)
i=1

9
i.e. sequences (x1 , x2 , . . . )of numbers such that |x1 |p + |x2 |p + . . . converges and ||x|| =
( ∞ p (1/p)
P
i=1 |xi | )
lp is a banach space.

1.2.7 Inner Product space

Inner product spaces are probably the most natural generalization of Euclidean space.Inner
product spaces are results of the question whether the dot product (a.b = α1 β1 +α2 β2 +α3 β3 )
and orthogonality (a.b = 0) can be generalized to arbitrary vector spaces or not. This spaces
have been the most useful spaces in practical applications in functional analysis and can be
defined as follows:

Definition- Let H be a vector space over K. An inner product on H is a mapping

<, >: H × H → K satisfying the following conditions:

For all x,y,z ∈H and α, β ∈ K

I1) ⟨x, x⟩ ⩾ 0
I2) ⟨x, x⟩ = 0 ⇔ x = 0
I3) ⟨x, y⟩ = ⟨y,¯x⟩ , where ⟨y,¯x⟩ is the complex conjugate if K = C
If K = R then ⟨x, y⟩ = ⟨y, x⟩
I4) ⟨x + y, z⟩ = ⟨x, z⟩ + ⟨y, z⟩
I5) ⟨αx, y⟩ = α⟨x, y⟩
The pair (H, ⟨, ⟩) is called an inner product space, i.e. an inner product space is a vector
space H with an inner product defined on H.

Example:
• Euclidean space Rn is an inner product space defined by-

⟨x, y⟩ = x1 y1 + x2 y2 + · · · + xn yn

where x = (xi ) = (x1 , x2 , . . . , xn ) and y = (yi ) = (y1 , y2 , . . . , yn )

Note:
• All inner product spaces are normed spaces as, an inner product on H defines a norm on
X given by p
||x|| = ⟨x, x⟩

10
, where x ⩾ 0
• A metric on H is given by
p
d(x, y) = ||x − y|| = ⟨x − y, x − y⟩
• But not all normed spaces are inner product space.
The space lp (1 ⩽ ∞ with p ̸= 2 is not an inner product space.

The space C[a, b] is not an inner product space.

1.2.8 Hilbert Space

Hilbert spaces were named after David Hilbert who studied them in the context of integral
equations.

Definition: A Hilbert space is a complete inner product space i.e. complete in the norm
induced by the inner product.
Examples:
Euclidean space Rn are Hilbert space.

The space X = C[a, b] of all scalar valued continuous functions on [a, b] is an inner product
space and Hilbert space defined by
Z b
⟨x, y⟩ = ¯
x(t)y(t)dt
a
and (1/2)
Z b
(1/2)) 2
⟨x, x⟩ = |x(t) |dt
a

Note:
• All Hilbert spaces are Banach spaces but the converse is not necessarily true.
• If x and y are two vectors in an inner product X, then there exists
a) Parallelogram law: ||x + y||2 + ||x − y||2 = 2(||x||2 + ||y||2 )
b) Polarization identity: ||x + y||2 − ||x − y||2 + i(||x + iy||2 − ||x − iy||2 ) = 4⟨x, y⟩

• Let p ≧ 1 be a fixed real number.Each element in the space lp is a sequence x = (ξi ) =

(ξ1 , ξ2 , . . . ) of numbers such that |ξ1 |p + |ξ2 |p + . . . converges, thus
∞
X
|ξi |p < ∞
i=1

11
(p ≧ 1 fixed) and the metric defined by

∞
!(1/p)
X
d(x, y) = |ξi − ηi |p
i=1

|ηi |p < ∞. From this we can derive the following-

P
where y = (ηi ) and

1.2.9 Holder inequality

 !(1/p)  !(1/q)
∞
X ∞
X ∞
X
|ξi ηi | ≦  |ξk |p  |ηm |q
i=1 k=1 m=1

1.2.10 Minkowski inequality

 !(1/p)   !(1/p)  !(1/p)
∞
X ∞
X ∞
X
 |ξi + ηi |p ≦ |ξk |p + |ηm |p
i=1 k=1 m=1

1.2.11 Cauchy Schwarz inequality

v v
∞
X
u∞
X
u ∞
u uX
|ξi ηi | ≦ t 2
|ξk | t |ηm |2
i=1 k=1 m=1

12
 2 Chapter
Linear operators

2.1 Operator

In calculus we are familiar with the real line R and real valued functions on R. In functional
analysis we consider more general spaces such as normed spaces, metric spaces and mappings
of these spaces.

In the case of vector spaces and in particular,normed spaces,a mapping from one normed
space to another is called an operator.

2.2 Linear operator

A linear operator T is an operator such that

i) The domain D(T)of T is a vector space and the range R(T) of T lies in a vector space
over the same field.
ii) For all x, y ∈ D(T ) and scalar α belongs to the field,

T (x + y) = T (x) + T (y)

T (αx) = αT (x)

or it can also be written as-

If X and Y are vector spaces over the field K(RorC), then T is an operator from X into Y
i.e. T : X → Y satisfying the following:

T (x + y) = T (x) + T (y)
T (αx) = αT (x)
or
T (αx + βy) = αT (x) + βT (y)
where x, y ∈ X and α, β ∈ K(RorC)
From above we can draw a result that linear operator T is a homomorphism of a vector space
into another vector space.

13
Examples:
•Identity Operator : The identity operator Ix : X → x is defined by Ix (x) = x for all
x∈X
•Zero operator : The zero operator 0 : X → Y is defined as 0(x) = 0 for all x ∈ X
•Differentiation : Let X be the vector space of all polynomials on [a, b]. We may define
a linear operator T on X by

T x(t) = t́
for every x ∈ X, where the prime denotes differentiation with respect to t. This oper-
ator T maps X onto itself.
•Integration : A linear operator T from C[a, b] into itself can be defined by
Z b
T x(t) = x(τ )dτ
a
t ∈ [a, b]

2.2.1 Kernel of T

The kernel of T denoted by ker(T) is the subspace T(−1) (0) of X.

A linear operator T is injective if and only if Ker(T)=0.

2.2.2 Rank of a linear operator

The rank if a linear operator is the dimension of its range.Thus a finite rank operator is a
linear operator with a finite dimensional range.

Theorem: Let T be a linear operator,then-

a) The range R(T ) is a vector space.
b) If dim D(T ) = n < ∞, then dimR(T ) ⩽ n, where D(T ) is the domain.
c) The null space N (T ) is a vector space.

Inverse operator Theorem: Let X and Y be two vector spaces both real or complex. Let
T : D(T ) → R(T ) be
a linear operator with domain D(T ) ⊂ X and range R(T ) ⊂ Y . Then-
a)The inverse T (−1) : R(T ) → D(T ) exists if and only if
Tx = 0 ⇒ x = 0

b) If T (−1) exists, it is a linear operator.

14
c) If dimD(T ) = n < ∞ and T (−1) exists, then dimR(T ) =dimD(T ).

Inverse of product Theorem: Let T : X → Y and S : Y → Z be bijective linear

operators, where X,Y,Z are vector spaces. Then the inverse

(ST )(−1) = T (−1) S (−1)

2.3 Bounded Linear Operator

Let X and Y be two normed spaces and T : D(T ) → Y be a linear operator T is said to be
bounded if there is a real number c such that for all x ∈ D(T )

||T x|| ⩽ c||x||

This shows that bounded linear operators maps bounded sets in Y.

Lemma(Norm): Let T be a bounded linear operator. Then-
An alternative formula for the norm of T is-

||T || = sup ||T x||

x∈D(T )

Examples of bounded linear operator

Identity operator: The Identity operator I : X → X on a normed space X ̸= (0) is
bounded and has a norm ||I|| = 1
Zero operator: The zero operator 0 : X → Y on a normed space X is bounded and has a
norm ||0|| = 0.
Integral Operator: We can define an integral operator T : C[0, 1] → C[0, 1] by y = T x
where Z 1
y(t) = k(t, ξ)x(ξ)dξ
0
where k is the kernel of T and the operator T here is bounded.
Theorem(Finite dimension): If a normed space X is finite dimensional, then every linear
operator on X is bounded.

2.4 Continuous Linear Operator

Let T : D(T ) → R(T ) be any operator whereD(T ) ⊂ X and R(T ) ⊂ Y and X and Y are
normed spaces. The operator T is continuous at a point (x0 ) ∈ D(T ) if for every ϵ > 0 there
is a δ > 0 such that

15
||T x − T x0 || < ϵ ∀x ∈ D(T ) satisfying ||x − x0 || < δ
T is continuous if T is continuous at everyx ∈ D(T ).

Continuity and Boundedness Theorem: Let T : D(T ) → Y be a linear operator, where

D(T ) ⊂ X and X,Y are normed spaces. Then-
(a) T is continuous if and only if T is bounded.
(b) If T is continuous at a single point, it is continuous.

Note:
A continuous linear operator is always a bounded linear operator but in general, a bounded
linear operator need not be continuous.

Continuity and Null space: Let T be a bounded linear operator. Then-

a) xn → x [where xn , x ∈ D(T )] ⇒ T xn → T x
b) The null space N (T ) is closed.

Equal operators: Two operators T1 and T2 are defined to be equal if they have the same
domain
D(T1 ) = D(T2 )
and if
T x1 = T x2
for all x ∈ D(T1 ) = D(T2 ). This can be written as T1 = T2

Extension of an Operator:
 An  extension of T to a set M ⊃ D(T ) is an operator-
T̃ : M → Y defined by T̃ |(D(T )) = T ,
that is T̃ x = T x for all x ∈ D(T )

• If D(T ) is a proper subset of M then a given T has many extensions.

Restriction of an operator: The restriction of an operator T : D(T ) → Y to a subset

B ⊂ D(T ) is denoted by T |B and is defined by
T |B : B → Y and T |B x = T x for all x ∈ B.

16
2.5 Linear Functionals

A functional is an operator whose range lies on the real line R or in the complex plane C.
We denote functionals by lowercase letters f,g,h,... and so on.

Definition: A linear functional f is a linear operator with domain in a vector space X and
range in the scalar field K of X, thus,

f : D(f ) → K

where K = R if X is real and K = C if X is complex.

Example:
•An important functional on C[a, b] is obtained if we choose a fixed t0 ∈ J = [a, b] and the
set f (x1 ) = x(t0 ) where x ∈ C[a, b]. f1 is linear, bounded and has norm ||f1 || = 1.
•The norm |||| : X → R on a normed space (X, || • || is a functional on X which is not linear.

2.5.1 Bounded linear functional

A bounded linear functional f is a bounded linear operator with range in the scalar field of
the normed space X in which the domain D(f ) lies. Thus there exists a real number c such
that for all x ∈ D(f )
|f (x)| ⩽ c||x||

Continuity and boundedness Theorem: A linear functional f with domain D(f ) in a

normed space is continuous if and only if f is bounded.

2.6 Normed spaces of Operators

We have already studied the concept of operators and have witnessed their importance. Here
we take two normed spaces X and Y and consider the set B(X, Y ), consisting of all bounded
linear operators from X into Y i.e. each such operator is defined on all of X and its range
lies in Y. Then we can show that B(X, Y ) is itself a normed space.

17
2.7 Dual Space

Let X be a normed space over K. The linear space ′X of all continuous linear functionals
on X with the norm given by-

|f (x)|

||f || = supx∈X = sup(||x||=1) |f (x)|
||x||
is called the normed dual or simply dual of X and is denoted by X́

Example: • The dual space of Rn is Rn .

• The dual space of l1 is l∞ .

Theorem: The dual space X́ of a normed space X is a Banach space.

2.8 Isomorphism of norm

Let T is a linear operator from a normed space X into a normed space Y. Then T is an
isomorphism or normed space isomorphism into Y if it is one-one and onto i.e. bijective and
continuous and its inverse mapping T ( − 1) is continuous on the range of T.

•The operator T is an isometric isomorphism or linear isometry if

||T x|| = ||x||

where x ∈ X.

• The space X is embedded in Y if there is an isomorphism from X to Y and is isometrically

embedded in Y if there is an isometric isomorphism from X into Y.

• If X and Y are isometric i.e. there exists an isomorphism from X onto Y then it is denoted
by
X∼=Y

Result: Let T be a linear operator from a normed space X into a normed space Y. Then-
a) The operator T is an isomorphism iff there are positive constants r and s such that

s||x|| ⩽ ||T x|| ⩽ t||x||

18
for x ∈ X
b) If T is an isometric isomorphism, then T is an isomorphism.
c) If X is a Banach space and T is an isomorphism, then T(x) is a Banach space.

2.9 Some Important Theorems

2.9.1 Hahn-Banach Theorem(Extension of linear functionals)

Let X be a vector space and p a sublinear functional on X. Furthermore, let f be a linear

functional which is defined on a subspace Zof X and satisfies

f (x) ≦ p(x)

for all x ∈ Z.
Then f has a linear extension f˜ from Z to X satisfying

f˜ ≦ p(x)

for all x ∈ X, that is f˜ is a linear functional on X satisfies the above equation on X and
˜ = f (x) for every x ∈ Z.
f (x)

2.9.2 Hahn-Banach Theorem(Generalized)

Let X be a real or complex vector space and p be a real valued functional on X which is
subadditive, that is, for all x, y ∈ X,

p(x + y) ≦ p(x) + p(y)

and for every scalar α satisfies

p(αx) = |α|p(x)
. Furthermore, let f be a linear functional which is defined on a subspace Z of X and satisfies

|f (x)| ≦ p(x)

for all x ∈ Z.
Then f has a linear extension f˜ from Z to X satisfying
˜ ≦ p(x)
|f (x)|

for all x ∈ X.
Hahn-Banach Theorem(Normed spaces) Let f be a bounded linear functional on a subspace

19
Z of a normed space X. Then there exists a bounded linear functional f˜ on X which is an
extension of f to X and has the same norm,
 
||f˜||x = ||f ||z
where
||f˜||x = sup |f˜(x)|,
x∈X||x||=1

||f ||z = sup |f (x)|

x∈Z||x||=1

(and ||f ||z = 0 in the trivial case Z =0).

2.9.3 Uniform Boundedness Theorem

Let (Tn ) be a sequence of bounded linear operators Tn : X → Y from a Banach space X

into a normed space Y such that (||Tn x||) is bounded for every x ∈ X, say,
||Tn || ≦ cx
where n=1,2,3,.... and cx is a real number. Then the sequence of norms ||Tn || is bounded,i.e,
there is a c such that
||Tn || ≦ c

2.9.4 Open Mapping Theorem

Definition(Open mapping): Let Xand Y be metric spaces. Then T : D(T ) → Y with

domain D(T ) ⊂ X is called an open mapping if for every open set in D(T ) the image is an
open set in Y.

Theorem: A bounded linear operator T from a Banach space X onto a Banach space Y is
an open mapping. Hence if T is bijective, T ( − 1) is continuous and thus bounded.

2.9.5 Closed Graph Theorem

Definition (Closed linear operator Let X and Y be two normed spaces and T : D(T ) →
Y a linear operator with domain D(T ) ⊂ X. Then T is called a closed linear operator if its
graph
g(T ) = [(x, y) x ∈ D(T ), y = T x]

20
is closed in the normed space X ∗ Y, where the two algebraic operations of a vector space
in X ∗ Y are defined as usual, i.e

(x1 , y1 ) + (x2 , y2 ) = (x1 + x2 , y1 + y2 )

α(x, y) = (αx, αy)

(α a scalar) and the norm on X ∗ Y is defined by

||(x, y)|| = ||x|| + ||y||.

Theorem Let X and Y be Banach spaces and T : D(T ) → Y a closed linear operator,
where D(T ) ⊂ X. Then if D(T ) is closed in X, the operator T is bounded.

21
 3 Chapter
2.
4.
1.
3.
Strong And Weak Convergence

3.1 Strong Convergence

In a normed space X the sequence {xn } is said to converge strongly if there exist an x ∈ X
such that

lim ∥ xn − x ∥= 0
n→∞

or,
lim xn = x
n→∞

Thus,we say that {xn } converges strongly to x i.e xn → x or x is strong limit of {xn }

3.2 Weak Convergence

In normed space X the sequence {xn } is said to be converge weakly if there is an x ∈ X such
that for every f ∈ X́,where X́ is dual space of a normed space X

lim f (xn ) = f (x)

n→∞

or,
xn → x
The element x is called the weak limit of {xn } and we say {xn } converges weakly to x. This
can also be written as -

|f (xn ) − f (x)| → 0 as n → ∞∀f ∈ X ∗

Weak convergence means convergence of sequence of scalers.

Theorem Let {xn } be a weakly convergent sequence in a normed space X, say xn → x.Then

(a) The weak limit x of (xn ) is unique.

Proof: Let xn → x and xn → y.

From definition, f (xn ) → f (x) , f (xn ) → f (y)

22
Since,{f (xn )} is a sequence of numbers,its limit is unique. Hence, f(x) = f(y) i.e, for every
f ∈ X́ we have,
f (x) − f (y) = f (x − y) = 0
⇒x−y =0
⇒x=y

(b) Every subsequence of (xn ) converges weakly to x.

Proof: Since from the first part {f (xn )} is a convergent sequence of numbers,so every sub-
sequence of {f (xn )} converges and possess the same limit as the sequence.

(c) The sequence (∥ xn ∥) is bounded.

proof: We know, every convergent sequence is bounded.Since, {f (xn )} is a convergent se-

quence of numbers and so it is bounded. say,

|f (xn )| ≤ cf , ∀n

,where cf is a constant depending on f. Using the canonical mapping,

C : X → X ′′

We can define gn ∈ X ′′ by g(f ) = f (xn ), f ∈ X́

Then, ∀n
|gn (f )| = |f (xn )| ≤ cf
, i.e, sequence (|gn (f )|) is bounded for every f ∈ X́

Since X́ is complete [By uniform boundness theorem ]

This implies, (∥ gn ∥) is bounded.

Now, ∥ gn ∥=∥ xn ∥

⇒ The sequence (∥ xn ∥) is bounded.

Theorem Let {xn } be a sequence in a normed space X.Then,

(a) Strong convergence implies weak convergence with the same limit,but the converse is not
generally true.

Proof: By definition,
xn → x
⇒ lim ∥ xn − x ∥= 0
n→∞

⇒ |f (xn ) − f (x)| = |f (xn − x)| ≤∥ f ∥∥ xn − x ∥→ 0, ∀f ∈ X́

23
 ⇒ xn → x[sincef (xn ) → f (x)]
The converse part can be proved using an orthonormal sequence en in a Hilbert space H.
Every f ∈ H́ has a Riesz representation f (x) =< x, z >.

Hence, f (en ) =< en , z >

Now, ∞ 2 2
P
n=1 | < en , z > | ≤∥ z ∥ (By Bessel’s inequality)

Thus, the series ∞ 2

P
n=1 | < en , z > | converges and its terms must approach zero as n → ∞

This implies, f (en ) =< en , z >→ 0

Since, f ∈ H́ was arbitrary

we see en → 0

But,(en ) doesn’t converge strongly because ∥ em − en ∥2 =< em − en , em − en >= 2

Hence,the Converse is not necessarily true.

(b)If dimX < ∞,then weak convergence ⇒strong convergence.

Proof: Let xn → x and dimX = k

Let, {e1 , e2 , . . . en } be any basis for X and say

(n) (n)
xn = α1 e1 + . . . αk ek

x = α1 e1 + . . . αk ek

By assumption, f (xn ) → f (x) for every f ∈ X́

We define, f1 , f2 , . . . , fk by
fj (ej ) = 1, fj (em ) = 0, m ̸= j
Then,
(n)
fj (xm ) = αj , fj (x) = αj

Hence,
(n)
fj (xm ) → fj (x) ⇒ αj → αj
Therefore,
k
X k
X
(n) (n)
∥ xn − x ∥=∥ (αj − αj )ej ∥≤ |αj − αj | ∥ ej ∥→ 0
j=1 j=1

as n → ∞.Thus,{xn } converges strongly to x. Hence, in a finite dimensional normed linear

space strong convergence and weak convergence are equivalent.

24
3.3 Convergence in Hilbert space

In a hilbert space xn → x if and only if < xn , z >→ < x, z > for all z in the space.

Proof: Given, xn → x ⇐⇒ f (xn ) → f (x)

Now,the Riesz’s theorem states that every bounded linear functional f on a hilbert space H
can be represented in terms of an inner product, namely, f (x) =< x, z > where z depends
on f and is uniquely determined by f and has a norm ∥ z ∥=∥ f ∥

⇐⇒< xn , z >→ < x, z >∀z ∈ H

∗ Thus,it can be stated that A sequence < xn > in a Hilbert space H is said to be weakly
convergent for x ∈ H,i.e, xn → x in H if {< xn , z >} → < x, z >∀z ∈ H

Result: In a normed space X we have xn → x if and only if

(a)The sequence (∥ xn ∥ is bounded.

(b)For every element f of a total subset M ⊂ X ∗ ,we have f (xn ) → f (x)

The proof of the above results are trivial.

3.4 Convergence of Sequence of Operators

Let X and Y be normed spaces.A sequence (Tn ) of operators Tn ∈ B(X, Y ) is said to be

∗ Uniformly operator convergent - if (Tn ) converges in the norm on B(X,Y).

∗ Strongly operator convergent-if (Tn ) converges strongly in Y for every x ∈ X

∗ Weakly operator convergent- if (Tn ) converges weakly in Y for every x ∈ X

This mathematically means that there is an operator T : X → Y such that

1)∥ Tn − T ∥→ 0

2)∥ Tn x − T x ∥→ 0 for all x ∈ X

3)|f (Tn x) − f (T x)| → 0 for all x ∈ X and f ∈ Ý respectively. T is called the uniform,
strong and weak operator limit of Tn respectively.

25
 4 Chapter
Spectrum

Spectrum is a part of spectral theory of linear operators. Spectral theory is one of the
main branches of modern functional analysis and its applications. Roughly speaking, it is
concerned with certain inverse operators, their general properties and their relations to the
original operators.It is very important for an understanding of operator themselves. In the
previous section an introductory information has been given of linear operators on normed
and inner product spaces.

We begin with finite dimensional spaces.Let T : X → X a linear operator where X is finite

dimensional. The operator T can be represented by matrices depending on the choice of
bases for X and so spectral theory of T is essentially matrix eigenvalue theory. To under-
stand spectrum we are to be aware of few basic definitions stated below:-

Eigenvalue and Eigenvector: An eigenvalue of a squarematrix A = (αij ) is a number λ

such that the equation
Ax = λx
has a solution x ̸= 0

Eigenvector of the matrix A is the vector x corresponding to the eigenvalue .

Eigenspace: The eigenvectors corresponding to the eigenvalue λ in

Ax = λx

and the zero vector form a vector subspace of X, where X is a normed space is called the
eigenspace of A corresponding to the eigenvalue λ.

4.1 Spectrum in finite dimensional normed space

The spectrum of an operator on a finite dimensional normed space is the set of all eigen
values.
Let T : X → X be a linear operator on a normed space X of finite dimension (say n).
Let e = (e1 , e2 , . . . , en ) be any basis for X and Te = (αij ) the matrix representing T with

26
respect to that basis. Then the eigenvalues of the matrix Te are called the eigenvalues of
the operator T , and the set of all eigenvalues of T is called the spectrum of T denoted by σ(T ).

The above statement is justified by-

Theorem(Eigenvalues of an operator): All matrices representing a given linear operator

T : X → X on a finite dimensional normed space X relative to various bases for X have the
same eigenvalues.

4.1.1 Resolvent set:

The complement of spectrum i.e. if σ(T ) is the spectrum of the operator T then, ρ(T ) =
C − σ(T ) is called the resolvent set of T .

Existence Theorem: A linear operator on a finite dimensional complex normed space

X ̸= 0 has atleast one eigenvalue.

4.2 Spectrum in infinite dimensional normed space

In this section we consider normed spaces of any dimension.

Few required terms and definition-

Let X ̸= (0) be a complex normed space and T : D(T ) → X a linear operator with domain
D(T ) ⊂ X. With T we associate the operator

Tλ = T − λI

where λ is a complex number and I is the identity operator on D(T ). If Tλ has an inverse,
we denote it by
(−1)
Rλ (T ) = Tλ = (T − λI)(−1)
and call it the resolvent operator of T or simply the resolvent of T .

Regular value A regular value λ of T ,where T is a linear operator T : D(T ) → X and λ

is a complex number such that

27
• Rλ (T ) exists.
• Rλ (T ) is bounded.
• Rλ (T ) is defined on a set which is dense in X.

4.2.1 Resolvent Set:

The resolvent set ρ(T )of T is set of all regular values λ of T .

Definition (Spectrum): The complement of the resolvent set σ(T ) = C − ρ(T ) in the
complex plane C is called the spectrum of T , and λ ∈ σ(T ) is called a spectral value of T .

The spectrum σ(T ) is partitioned into three disjoint sets as follows-

4.2.2 The Point Spectrum:

The point spectrum σp (T ) is the set such that the resolvent Rλ (T ) does not exist i.e. Tλ is
not invertible and λ ∈ σp (T ) is called an eigenvalue of T .

4.2.3 The Continuous Spectrum

The continuous spectrum is the such that Rλ (T ) exists and is defined on a set dense in X
but Rλ (T ) is not bounded. This set is denoted by σc (T ).

4.2.4 The Residual Spectrum

(
The residual spectrum is the set such that Rλ (T ) exists but the domain of Rλ (T ) = Tλ − 1)
is not dense in X.

28
Theorem(Inverse): Let B(X, X), where X is a Banach space. If ||T|| < 1, then
∞
X
(I − T)(−1) = Tj = I + T + T2 + . . .
j=0

Theorem(Spectrum): The spectrum σ(T ) of a bounded linear operator T : X → X on a

complex Banach space X is compact and lies in the disk given by

|λ| ≦ ||T ||

Proof: Let λ ̸= 0 and K = 1/λ.

From the inverse theorem we obtain the representation
∞ ∞
(−1) 1 1X 1X
Rλ = (T − λI) = − (I − KT)(−1) = − (KT)j = − (1/λT)j
λ λ j=0 λ j=0

where the series converges for all λ such that

1 ||T||
∥ T∥ =
λ |λ|

i.e. |λ| > ||T||.

Theorem(Spectrum Closed): The resolvent set ρ(T) of a bounded linear operator T on

a complex Banach space X is open, hence the spectrum σ(T) is closed.

4.2.5 Relation to eigenvalues

If λ is an eigenvalue of T, then the operator T − λI is not one-one and therefore its inverse
is not defined. However, the inverse statement is not true i.e, the operator T − λI may not
have an inverse, even if λ is not an eigenvalue. Thus, the spectrum of an operator always
contain all its eigenvalues, but is not limited to them.
In fact in finite dimensional spaces the spectrum is the set of all eigenvalues.

4.2.6 Spectrum of a Bounded Operator

Let T be a bounded linear operator acting on a Banach space X over the complex scalar
field C and I be the identity operator on X. The spectrum of T is the set of all λ ∈ C for

29
which the operator T − λI does not have an inverse, that is a bounded linear operator.

•The spectrum of a bounded operator T is always closed, bounded and non empty subset
of the complex plane.

•If the spectrum were empty, then the resolvent function R(λ) = (T − λI)(−1) , λ ∈ C would
be defined everywhere on the complex plane and bounded.

4.2.7 Spectrum of an unbounded operator

The definition of spectrum to unbounded operators can be extended on a Banach space X.

Let X be a Banach space and T : X → X be a linear operator defined on the domain
D(T ) ⊂ X. A complex number λ is then in the spectrum if λ is not in the resolvent set.

• The spectrum of an unbounded operator is on general a closed, possibly empty,subset of

the complex plane.If the operator T is not closed then σ(T) = C.

30
 5 Conclusion

In this project, we first studied the very basics of functional analysis and some of its basic
important results. Then we further continued the study of linear operators in spaces, their
properties and results as well. Next we introduced a the topic of strong and weak conver-
gence of sequences in brief, followed by introductory portion of spectrum that forms the very
basis of spectral theory of linear operators.

Enormous number of papers has been released regarding the study of functional analysis
and operator theory as well, making it a vast topic with numerous applications in field
of mathematics.Though the boundedness of the project and time did not allow for further
exploration of the operators, but it has a wide range of scope of study and discovery to be
made.

31
 6 Bibliography

• KREYSZIG,Introductory Functional Analysis with Applications.

• Balmohan V Limaye, Functional Analysis.

•Theo Bühler and Dietmar A. Salamon, Functional analysis.

•https://en.m.wikipedia.org/wiki/Spectrum( f unctionala nalysis)

• https://www.math.uzh.ch

32