1 2 3,, - indian institute of technology · pdf file• it is very difficult to pinpoint...

176
December 16, 2003 Let us count 1, 2, 3,..., Prof. Inder K. Rana [email protected] Department of Mathematics Indian Institute of Technology Bombay, Powai Mumbai, 400076, India Inder K. Rana, IIT Bombay – p. 1/30

Upload: doanphuc

Post on 08-Mar-2018

214 views

Category:

Documents


1 download

TRANSCRIPT

December16, 2003

Let us count

1, 2, 3, . . . ,∞

Prof. Inder K. Rana

[email protected]

Department of Mathematics

Indian Institute of Technology Bombay, Powai

Mumbai, 400076, India

Inder K. Rana, IIT Bombay – p. 1/30

December16, 2003

Origin of counting and number symbols

• It is very difficult to pinpoint the origin of numbers and theprocess of counting.

Inder K. Rana, IIT Bombay – p. 2/30

December16, 2003

Origin of counting and number symbols

• It is very difficult to pinpoint the origin of numbers and theprocess of counting.

• Of course, this must have been a process of gradualawareness developing over centuries, rather than a discovery.

Inder K. Rana, IIT Bombay – p. 2/30

December16, 2003

Origin of counting and number symbols

• It is very difficult to pinpoint the origin of numbers and theprocess of counting.

• Of course, this must have been a process of gradualawareness developing over centuries, rather than a discovery.

• The need to express these similarities or differences insymbols − first orally and then written − must have given riseto the process of counting and to number symbols.

Inder K. Rana, IIT Bombay – p. 2/30

December16, 2003

Origin of counting and number symbols

• It is very difficult to pinpoint the origin of numbers and theprocess of counting.

• Of course, this must have been a process of gradualawareness developing over centuries, rather than a discovery.

• The need to express these similarities or differences insymbols − first orally and then written − must have given riseto the process of counting and to number symbols.

• Anthropologists have discoveries to their credit to claim thatthis process must have begun at least about 30,000 yearsback.

Inder K. Rana, IIT Bombay – p. 2/30

December16, 2003

Origin of counting and number symbols

Inder K. Rana, IIT Bombay – p. 3/30

December16, 2003

Origin of counting and number symbols

Inder K. Rana, IIT Bombay – p. 3/30

December16, 2003

Origin of counting and number symbols

Inder K. Rana, IIT Bombay – p. 3/30

December16, 2003

Origin of counting and number symbols

Inder K. Rana, IIT Bombay – p. 4/30

December16, 2003

Origin of counting and number symbols

Inder K. Rana, IIT Bombay – p. 4/30

December16, 2003

Origin of counting and number symbols

Inder K. Rana, IIT Bombay – p. 4/30

December16, 2003

Origin of counting and number symbols

• Around 1400 A.D, these symbols and the decimal place-valuesystem reached Europe and rest of the world via the Arabtraders.

Inder K. Rana, IIT Bombay – p. 4/30

December16, 2003

Number symbols

Evolution of number symbols

Inder K. Rana, IIT Bombay – p. 5/30

December16, 2003

Number symbols

• The current number symbols 1, 2, . . . , 9 and the decimalplace value system help us to represent all the numbers.

Inder K. Rana, IIT Bombay – p. 6/30

December16, 2003

Number symbols

• The current number symbols 1, 2, . . . , 9 and the decimalplace value system help us to represent all the numbers.

• These numbers are called natural numbers and we denotethe set of natural numbers by IN.

Inder K. Rana, IIT Bombay – p. 6/30

December16, 2003

Number symbols

• The current number symbols 1, 2, . . . , 9 and the decimalplace value system help us to represent all the numbers.

• These numbers are called natural numbers and we denotethe set of natural numbers by IN.

• As we know, every method of counting is a process ofassociating, with each object to be counted, a unique naturalnumber in a one-to-one manner.

Inder K. Rana, IIT Bombay – p. 6/30

December16, 2003

Number symbols

• The current number symbols 1, 2, . . . , 9 and the decimalplace value system help us to represent all the numbers.

• These numbers are called natural numbers and we denotethe set of natural numbers by IN.

• As we know, every method of counting is a process ofassociating, with each object to be counted, a unique naturalnumber in a one-to-one manner.

• The process of counting is based on the following:

Pigeon hole principle:

Inder K. Rana, IIT Bombay – p. 6/30

December16, 2003

Number symbols

• The current number symbols 1, 2, . . . , 9 and the decimalplace value system help us to represent all the numbers.

• These numbers are called natural numbers and we denotethe set of natural numbers by IN.

• As we know, every method of counting is a process ofassociating, with each object to be counted, a unique naturalnumber in a one-to-one manner.

• The process of counting is based on the following:

Pigeon hole principle:

1. If m pigeons are put in m holes then there is an empty holeif and only if there is a hole with more than one pigeon.

Inder K. Rana, IIT Bombay – p. 6/30

December16, 2003

Number symbols

• The current number symbols 1, 2, . . . , 9 and the decimalplace value system help us to represent all the numbers.

• These numbers are called natural numbers and we denotethe set of natural numbers by IN.

• As we know, every method of counting is a process ofassociating, with each object to be counted, a unique naturalnumber in a one-to-one manner.

• The process of counting is based on the following:

Pigeon hole principle:

1. If m pigeons are put in m holes then there is an empty holeif and only if there is a hole with more than one pigeon.

2. If m pigeons are put in n holes, m > n, then there is a holewith more than one pigeon.

Inder K. Rana, IIT Bombay – p. 6/30

December16, 2003

Pigeon Hole Principle

• Mathematically these can be stated as follows:

Inder K. Rana, IIT Bombay – p. 7/30

December16, 2003

Pigeon Hole Principle

• Mathematically these can be stated as follows:1. For n ∈ IN, every one-one map f : {1, 2, . . . , n} → {1, 2, . . . , n} is

also onto.

Inder K. Rana, IIT Bombay – p. 7/30

December16, 2003

Pigeon Hole Principle

• Mathematically these can be stated as follows:1. For n ∈ IN, every one-one map f : {1, 2, . . . , n} → {1, 2, . . . , n} is

also onto.

2. For n,m ∈ IN there exists a bijective mapΦ : {1, 2, . . . , n} → {1, 2, . . . ,m} if and only if n = m.

Inder K. Rana, IIT Bombay – p. 7/30

December16, 2003

Pigeon Hole Principle

• Mathematically these can be stated as follows:1. For n ∈ IN, every one-one map f : {1, 2, . . . , n} → {1, 2, . . . , n} is

also onto.

2. For n,m ∈ IN there exists a bijective mapΦ : {1, 2, . . . , n} → {1, 2, . . . ,m} if and only if n = m.

• Some applications of the Pigeon Hole Principle are as follows:

Inder K. Rana, IIT Bombay – p. 7/30

December16, 2003

Pigeon Hole Principle

• Mathematically these can be stated as follows:1. For n ∈ IN, every one-one map f : {1, 2, . . . , n} → {1, 2, . . . , n} is

also onto.

2. For n,m ∈ IN there exists a bijective mapΦ : {1, 2, . . . , n} → {1, 2, . . . ,m} if and only if n = m.

• Some applications of the Pigeon Hole Principle are as follows:1. Among any group of 8 persons, two of them have same day of the week

as birthday.

Inder K. Rana, IIT Bombay – p. 7/30

December16, 2003

Pigeon Hole Principle

• Mathematically these can be stated as follows:1. For n ∈ IN, every one-one map f : {1, 2, . . . , n} → {1, 2, . . . , n} is

also onto.

2. For n,m ∈ IN there exists a bijective mapΦ : {1, 2, . . . , n} → {1, 2, . . . ,m} if and only if n = m.

• Some applications of the Pigeon Hole Principle are as follows:1. Among any group of 8 persons, two of them have same day of the week

as birthday.

2. Among any N positive integers, there exists a pair whose difference isdivisible by N − 1.

Inder K. Rana, IIT Bombay – p. 7/30

December16, 2003

Application of PHP

• Proof of (2): Let a1, a2, . . . , aN be the given numbers.

Inder K. Rana, IIT Bombay – p. 8/30

December16, 2003

Application of PHP

• Proof of (2): Let a1, a2, . . . , aN be the given numbers.

For each ai, let ri be the remainder that results from dividingai by N − 1.

Inder K. Rana, IIT Bombay – p. 8/30

December16, 2003

Application of PHP

• Proof of (2): Let a1, a2, . . . , aN be the given numbers.

For each ai, let ri be the remainder that results from dividingai by N − 1.

Each ri can take one the values 0, 1, ..., N − 2.

Inder K. Rana, IIT Bombay – p. 8/30

December16, 2003

Application of PHP

• Proof of (2): Let a1, a2, . . . , aN be the given numbers.

For each ai, let ri be the remainder that results from dividingai by N − 1.

Each ri can take one the values 0, 1, ..., N − 2.

Thus there are N − 1 possible values for N remainders:r1, r2, . . . , rN .

Inder K. Rana, IIT Bombay – p. 8/30

December16, 2003

Application of PHP

• Proof of (2): Let a1, a2, . . . , aN be the given numbers.

For each ai, let ri be the remainder that results from dividingai by N − 1.

Each ri can take one the values 0, 1, ..., N − 2.

Thus there are N − 1 possible values for N remainders:r1, r2, . . . , rN .

Hence, by the pigeon hole principle, there must be two of ther′is that are the same,i,e., rj = rk for some pair j and k.

Inder K. Rana, IIT Bombay – p. 8/30

December16, 2003

Application of PHP

• Proof of (2): Let a1, a2, . . . , aN be the given numbers.

For each ai, let ri be the remainder that results from dividingai by N − 1.

Each ri can take one the values 0, 1, ..., N − 2.

Thus there are N − 1 possible values for N remainders:r1, r2, . . . , rN .

Hence, by the pigeon hole principle, there must be two of ther′is that are the same,i,e., rj = rk for some pair j and k.

But then, aj and ak have the same remainder when divided byN − 1, and so their difference aj − ak is divisible by N − 1.

Inder K. Rana, IIT Bombay – p. 8/30

December16, 2003

Finite sets

In view of the pigeon hole principle, we have the following:

Inder K. Rana, IIT Bombay – p. 9/30

December16, 2003

Finite sets

In view of the pigeon hole principle, we have the following:• Definition:

A set A is said to be countably finite if either A = ∅ or thereexists an n ∈ IN and a bijective map φ : A→ {1, 2, . . . , n}.

Inder K. Rana, IIT Bombay – p. 9/30

December16, 2003

Finite sets

In view of the pigeon hole principle, we have the following:• Definition:

A set A is said to be countably finite if either A = ∅ or thereexists an n ∈ IN and a bijective map φ : A→ {1, 2, . . . , n}.

Note that by the PHP, such a n is unique, and we say that Ahas n elements.

Inder K. Rana, IIT Bombay – p. 9/30

December16, 2003

Finite sets

In view of the pigeon hole principle, we have the following:• Definition:

A set A is said to be countably finite if either A = ∅ or thereexists an n ∈ IN and a bijective map φ : A→ {1, 2, . . . , n}.

Note that by the PHP, such a n is unique, and we say that Ahas n elements.

The following properties of finite sets are easy to check:

Inder K. Rana, IIT Bombay – p. 9/30

December16, 2003

Finite sets

In view of the pigeon hole principle, we have the following:• Definition:

A set A is said to be countably finite if either A = ∅ or thereexists an n ∈ IN and a bijective map φ : A→ {1, 2, . . . , n}.

Note that by the PHP, such a n is unique, and we say that Ahas n elements.

The following properties of finite sets are easy to check:• Theorem:

Let A and B be two sets. Then the following hold:

Inder K. Rana, IIT Bombay – p. 9/30

December16, 2003

Finite sets

In view of the pigeon hole principle, we have the following:• Definition:

A set A is said to be countably finite if either A = ∅ or thereexists an n ∈ IN and a bijective map φ : A→ {1, 2, . . . , n}.

Note that by the PHP, such a n is unique, and we say that Ahas n elements.

The following properties of finite sets are easy to check:• Theorem:

Let A and B be two sets. Then the following hold:

(i) If A is countably finite and B is countably finite, so also is A ∪B.

Inder K. Rana, IIT Bombay – p. 9/30

December16, 2003

Finite sets

In view of the pigeon hole principle, we have the following:• Definition:

A set A is said to be countably finite if either A = ∅ or thereexists an n ∈ IN and a bijective map φ : A→ {1, 2, . . . , n}.

Note that by the PHP, such a n is unique, and we say that Ahas n elements.

The following properties of finite sets are easy to check:• Theorem:

Let A and B be two sets. Then the following hold:

(i) If A is countably finite and B is countably finite, so also is A ∪B.

(ii) If A1, A2, . . . , An are finite sets, Then so is the set ∪ni=1Ai.

Inder K. Rana, IIT Bombay – p. 9/30

December16, 2003

Countably-infinite

At this point it is natural to ask the question:

How to analyze sets which are not finite?

Inder K. Rana, IIT Bombay – p. 10/30

December16, 2003

Countably-infinite

At this point it is natural to ask the question:

How to analyze sets which are not finite?

For example, IN, the set of all natural numbers is not finite!

Inder K. Rana, IIT Bombay – p. 10/30

December16, 2003

Countably-infinite

At this point it is natural to ask the question:

How to analyze sets which are not finite?

For example, IN, the set of all natural numbers is not finite!

It is natural to call such sets as infinite sets.

Inder K. Rana, IIT Bombay – p. 10/30

December16, 2003

Countably-infinite

At this point it is natural to ask the question:

How to analyze sets which are not finite?

For example, IN, the set of all natural numbers is not finite!

It is natural to call such sets as infinite sets.

The concepts of ‘infinite’ and ‘infinite set’ eludedmathematicians and philosophers over the centuries.

Inder K. Rana, IIT Bombay – p. 10/30

December16, 2003

Countably-infinite

At this point it is natural to ask the question:

How to analyze sets which are not finite?

For example, IN, the set of all natural numbers is not finite!

It is natural to call such sets as infinite sets.

The concepts of ‘infinite’ and ‘infinite set’ eludedmathematicians and philosophers over the centuries.

• The Greek mathematician, Pythagoras (∼ 585–500 B.C.),associated good and evil with the limited and the unlimited,respectively.

Inder K. Rana, IIT Bombay – p. 10/30

December16, 2003

Countably-infinite

At this point it is natural to ask the question:

How to analyze sets which are not finite?

For example, IN, the set of all natural numbers is not finite!

It is natural to call such sets as infinite sets.

The concepts of ‘infinite’ and ‘infinite set’ eludedmathematicians and philosophers over the centuries.

• The Greek mathematician, Pythagoras (∼ 585–500 B.C.),associated good and evil with the limited and the unlimited,respectively.

• “The infinite is imperfect, unfinished and therefore,unthinkable; it is formless and confused," said, Aristotle(384–322 B.C.).

Inder K. Rana, IIT Bombay – p. 10/30

December16, 2003

Countably-infinite

At this point it is natural to ask the question:

How to analyze sets which are not finite?

For example, IN, the set of all natural numbers is not finite!

It is natural to call such sets as infinite sets.

The concepts of ‘infinite’ and ‘infinite set’ eludedmathematicians and philosophers over the centuries.

• The Greek mathematician, Pythagoras (∼ 585–500 B.C.),associated good and evil with the limited and the unlimited,respectively.

• “The infinite is imperfect, unfinished and therefore,unthinkable; it is formless and confused," said, Aristotle(384–322 B.C.).

Inder K. Rana, IIT Bombay – p. 10/30

December16, 2003

Infinity

• The Roman Emperor and philosopher Marcus Aqarchus(121-180 A.D.) said, “Infinity is a fathomless gulf, into which allthings vanish".

Inder K. Rana, IIT Bombay – p. 11/30

December16, 2003

Infinity

• The Roman Emperor and philosopher Marcus Aqarchus(121-180 A.D.) said, “Infinity is a fathomless gulf, into which allthings vanish".

• English philosopher Thomas Hobbes (1588–1679) said,“When we say any thing is infinite, we signify only that we arenot able conceive the ends and bounds of the thing named".

Inder K. Rana, IIT Bombay – p. 11/30

December16, 2003

Infinity

• The Roman Emperor and philosopher Marcus Aqarchus(121-180 A.D.) said, “Infinity is a fathomless gulf, into which allthings vanish".

• English philosopher Thomas Hobbes (1588–1679) said,“When we say any thing is infinite, we signify only that we arenot able conceive the ends and bounds of the thing named".

• The German mathematician Carl Friedrich Gauss(1777–1855) said, “Infinity is only a figure of speech, meaninga limit to which certain ratios may approach as closely asdesired, when others are permitted to increase indefinitely".

Inder K. Rana, IIT Bombay – p. 11/30

December16, 2003

Infinity

• The Roman Emperor and philosopher Marcus Aqarchus(121-180 A.D.) said, “Infinity is a fathomless gulf, into which allthings vanish".

• English philosopher Thomas Hobbes (1588–1679) said,“When we say any thing is infinite, we signify only that we arenot able conceive the ends and bounds of the thing named".

• The German mathematician Carl Friedrich Gauss(1777–1855) said, “Infinity is only a figure of speech, meaninga limit to which certain ratios may approach as closely asdesired, when others are permitted to increase indefinitely".

• Infinity is indeed strange and abstract.

Inder K. Rana, IIT Bombay – p. 11/30

December16, 2003

Infinity

• The Roman Emperor and philosopher Marcus Aqarchus(121-180 A.D.) said, “Infinity is a fathomless gulf, into which allthings vanish".

• English philosopher Thomas Hobbes (1588–1679) said,“When we say any thing is infinite, we signify only that we arenot able conceive the ends and bounds of the thing named".

• The German mathematician Carl Friedrich Gauss(1777–1855) said, “Infinity is only a figure of speech, meaninga limit to which certain ratios may approach as closely asdesired, when others are permitted to increase indefinitely".

• Infinity is indeed strange and abstract.• Let us look at some examples.

Inder K. Rana, IIT Bombay – p. 11/30

December16, 2003

Infinity

• The Roman Emperor and philosopher Marcus Aqarchus(121-180 A.D.) said, “Infinity is a fathomless gulf, into which allthings vanish".

• English philosopher Thomas Hobbes (1588–1679) said,“When we say any thing is infinite, we signify only that we arenot able conceive the ends and bounds of the thing named".

• The German mathematician Carl Friedrich Gauss(1777–1855) said, “Infinity is only a figure of speech, meaninga limit to which certain ratios may approach as closely asdesired, when others are permitted to increase indefinitely".

• Infinity is indeed strange and abstract.• Let us look at some examples.• Zeno’s Paradox:

Inder K. Rana, IIT Bombay – p. 11/30

December16, 2003

Infinity

• The Roman Emperor and philosopher Marcus Aqarchus(121-180 A.D.) said, “Infinity is a fathomless gulf, into which allthings vanish".

• English philosopher Thomas Hobbes (1588–1679) said,“When we say any thing is infinite, we signify only that we arenot able conceive the ends and bounds of the thing named".

• The German mathematician Carl Friedrich Gauss(1777–1855) said, “Infinity is only a figure of speech, meaninga limit to which certain ratios may approach as closely asdesired, when others are permitted to increase indefinitely".

• Infinity is indeed strange and abstract.• Let us look at some examples.• Zeno’s Paradox:

A man standing in a room can not walk to the wall he is facing.

Inder K. Rana, IIT Bombay – p. 11/30

December16, 2003

Zeno’s paradx

◦ ◦ ◦

1/2 1/4 1/8

6

Man → Wall

�������

Inder K. Rana, IIT Bombay – p. 12/30

December16, 2003

Zeno’s paradx

◦ ◦ ◦

1/2 1/4 1/8

6

Man → Wall

�������

Because, in order to do so, he would have to go half thedistance,

◦ ◦ ◦

1/2 1/4 1/8

6

Man →

Wall

�������

Inder K. Rana, IIT Bombay – p. 12/30

December16, 2003

Zeno’s paradx

◦ ◦ ◦

1/2 1/4 1/8

6

Man → Wall

�������

Because, in order to do so, he would have to go half thedistance,

◦ ◦ ◦

1/2 1/4 1/8

6

Man →

Wall

�������

then half the remaining distance,Inder K. Rana, IIT Bombay – p. 12/30

December16, 2003

Zeno’s paradox

• and then again half of what shall remain.

Inder K. Rana, IIT Bombay – p. 13/30

December16, 2003

Zeno’s paradox

• and then again half of what shall remain.

This process can always be continued and it can never end.

Inder K. Rana, IIT Bombay – p. 13/30

December16, 2003

Zeno’s paradox

• and then again half of what shall remain.

This process can always be continued and it can never end.

However, physically the man walks to the wall in finite time.

Inder K. Rana, IIT Bombay – p. 13/30

December16, 2003

Zeno’s paradox

• and then again half of what shall remain.

This process can always be continued and it can never end.

However, physically the man walks to the wall in finite time.• The paradox is that there are infinite stages and how they can

be covered in finite time?

Inder K. Rana, IIT Bombay – p. 13/30

December16, 2003

Zeno’s paradox

• and then again half of what shall remain.

This process can always be continued and it can never end.

However, physically the man walks to the wall in finite time.• The paradox is that there are infinite stages and how they can

be covered in finite time?• Explanation: Let the man walk with a constant speed and takea minutes to cover the first half of the distance.

Inder K. Rana, IIT Bombay – p. 13/30

December16, 2003

Zeno’s paradox

• and then again half of what shall remain.

This process can always be continued and it can never end.

However, physically the man walks to the wall in finite time.• The paradox is that there are infinite stages and how they can

be covered in finite time?• Explanation: Let the man walk with a constant speed and takea minutes to cover the first half of the distance.

The next half will be covered in a/2 minutes, the half of theremaining half in a/4 minutes, and so on.

Inder K. Rana, IIT Bombay – p. 13/30

December16, 2003

Zeno’s paradox

• and then again half of what shall remain.

This process can always be continued and it can never end.

However, physically the man walks to the wall in finite time.• The paradox is that there are infinite stages and how they can

be covered in finite time?• Explanation: Let the man walk with a constant speed and takea minutes to cover the first half of the distance.

The next half will be covered in a/2 minutes, the half of theremaining half in a/4 minutes, and so on.

The time consumed at the nth stage will be

Tn = a+a

2+

a

22+ · · · +

a

2n−1,

which is ’convergent’ to a.

Inder K. Rana, IIT Bombay – p. 13/30

December16, 2003

Zeno’s paradox

• and then again half of what shall remain.

This process can always be continued and it can never end.

However, physically the man walks to the wall in finite time.• The paradox is that there are infinite stages and how they can

be covered in finite time?• Explanation: Let the man walk with a constant speed and takea minutes to cover the first half of the distance.

The next half will be covered in a/2 minutes, the half of theremaining half in a/4 minutes, and so on.

The time consumed at the nth stage will be

Tn = a+a

2+

a

22+ · · · +

a

2n−1,

which is ’convergent’ to a.

Inder K. Rana, IIT Bombay – p. 13/30

December16, 2003

Hotel Hilbert

• Another illustration of the mysteries of the infinite is thefollowing :

Inder K. Rana, IIT Bombay – p. 14/30

December16, 2003

Hotel Hilbert

• Another illustration of the mysteries of the infinite is thefollowing :

• Problem:To construct a hotel which always has vacant rooms!

Inder K. Rana, IIT Bombay – p. 14/30

December16, 2003

Hotel Hilbert

• Another illustration of the mysteries of the infinite is thefollowing :

• Problem:To construct a hotel which always has vacant rooms!

Solution:Construct a hotel with infinite number of rooms.

Inder K. Rana, IIT Bombay – p. 14/30

December16, 2003

Hotel Hilbert

• Another illustration of the mysteries of the infinite is thefollowing :

• Problem:To construct a hotel which always has vacant rooms!

Solution:Construct a hotel with infinite number of rooms.

The solution was implemented and the new hotel was called:

Inder K. Rana, IIT Bombay – p. 14/30

December16, 2003

Hotel Hilbert

• Another illustration of the mysteries of the infinite is thefollowing :

• Problem:To construct a hotel which always has vacant rooms!

Solution:Construct a hotel with infinite number of rooms.

The solution was implemented and the new hotel was called:

... HHHHHHHHHOTEL HILBERT TTTTTTTT ...

Inder K. Rana, IIT Bombay – p. 14/30

December16, 2003

Hotel Hilbert

• Another illustration of the mysteries of the infinite is thefollowing :

• Problem:To construct a hotel which always has vacant rooms!

Solution:Construct a hotel with infinite number of rooms.

The solution was implemented and the new hotel was called:

... HHHHHHHHHOTEL HILBERT TTTTTTTT ...

The rooms in the hotel were along one big long corridor,

Inder K. Rana, IIT Bombay – p. 14/30

December16, 2003

Hotel Hilbert

• Another illustration of the mysteries of the infinite is thefollowing :

• Problem:To construct a hotel which always has vacant rooms!

Solution:Construct a hotel with infinite number of rooms.

The solution was implemented and the new hotel was called:

... HHHHHHHHHOTEL HILBERT TTTTTTTT ...

The rooms in the hotel were along one big long corridor,

with all the odd numbered rooms 1, 3, 5, 7, . . . , on the left handside of the hall and at the right side were all the evennumbered rooms 2, 4, 6, 8, . . ..

Inder K. Rana, IIT Bombay – p. 14/30

December16, 2003

Hotel Hilbert

• Another illustration of the mysteries of the infinite is thefollowing :

• Problem:To construct a hotel which always has vacant rooms!

Solution:Construct a hotel with infinite number of rooms.

The solution was implemented and the new hotel was called:

... HHHHHHHHHOTEL HILBERT TTTTTTTT ...

The rooms in the hotel were along one big long corridor,

with all the odd numbered rooms 1, 3, 5, 7, . . . , on the left handside of the hall and at the right side were all the evennumbered rooms 2, 4, 6, 8, . . ..

Everybody was happy that this hotel had accommodation forevery visitor.

Inder K. Rana, IIT Bombay – p. 14/30

December16, 2003

Hotel Hilbert

However, one day the most unbelievable thing happened anda problem came up:

Inder K. Rana, IIT Bombay – p. 15/30

December16, 2003

Hotel Hilbert

However, one day the most unbelievable thing happened anda problem came up:

• New Problem:One day the receptionist wanted to accommodate a newvisitor, but observed that the hotel was full!!

Inder K. Rana, IIT Bombay – p. 15/30

December16, 2003

Hotel Hilbert

However, one day the most unbelievable thing happened anda problem came up:

• New Problem:One day the receptionist wanted to accommodate a newvisitor, but observed that the hotel was full!!

• The perplexed receptionist went to the manager of the hotel.

Inder K. Rana, IIT Bombay – p. 15/30

December16, 2003

Hotel Hilbert

However, one day the most unbelievable thing happened anda problem came up:

• New Problem:One day the receptionist wanted to accommodate a newvisitor, but observed that the hotel was full!!

• The perplexed receptionist went to the manager of the hotel.• The manager thought for a while, and solved the problem as

follows:

Inder K. Rana, IIT Bombay – p. 15/30

December16, 2003

Hotel Hilbert

However, one day the most unbelievable thing happened anda problem came up:

• New Problem:One day the receptionist wanted to accommodate a newvisitor, but observed that the hotel was full!!

• The perplexed receptionist went to the manager of the hotel.• The manager thought for a while, and solved the problem as

follows:• Solution: Move occupant of Room1 to Room2, move occupant

of Room2 to Room3, and in general

Inder K. Rana, IIT Bombay – p. 15/30

December16, 2003

Hotel Hilbert

However, one day the most unbelievable thing happened anda problem came up:

• New Problem:One day the receptionist wanted to accommodate a newvisitor, but observed that the hotel was full!!

• The perplexed receptionist went to the manager of the hotel.• The manager thought for a while, and solved the problem as

follows:• Solution: Move occupant of Room1 to Room2, move occupant

of Room2 to Room3, and in general

move occupant of RoomN to RoomN + 1.

Inder K. Rana, IIT Bombay – p. 15/30

December16, 2003

Hotel Hilbert

However, one day the most unbelievable thing happened anda problem came up:

• New Problem:One day the receptionist wanted to accommodate a newvisitor, but observed that the hotel was full!!

• The perplexed receptionist went to the manager of the hotel.• The manager thought for a while, and solved the problem as

follows:• Solution: Move occupant of Room1 to Room2, move occupant

of Room2 to Room3, and in general

move occupant of RoomN to RoomN + 1.

This will make the Room1 available for the new guest.

Inder K. Rana, IIT Bombay – p. 15/30

December16, 2003

Hotel Hilbert

However, one day the most unbelievable thing happened anda problem came up:

• New Problem:One day the receptionist wanted to accommodate a newvisitor, but observed that the hotel was full!!

• The perplexed receptionist went to the manager of the hotel.• The manager thought for a while, and solved the problem as

follows:• Solution: Move occupant of Room1 to Room2, move occupant

of Room2 to Room3, and in general

move occupant of RoomN to RoomN + 1.

This will make the Room1 available for the new guest.

Business went smoothly for the hotel till one day, thereceptionist came back to the manager with yet anotherproblem:

Inder K. Rana, IIT Bombay – p. 15/30

December16, 2003

Hotel Hilbert

• Yet another problem:Infinite number of new visitors had arrived asking foraccommodation, and the hotel was full !!!

Inder K. Rana, IIT Bombay – p. 16/30

December16, 2003

Hotel Hilbert

• Yet another problem:Infinite number of new visitors had arrived asking foraccommodation, and the hotel was full !!!

The manger was out of his wits to solve the problem. Luckily,a mathematician friend of his was in his office, and heproposed a solution the manager.

Inder K. Rana, IIT Bombay – p. 16/30

December16, 2003

Hotel Hilbert

• Yet another problem:Infinite number of new visitors had arrived asking foraccommodation, and the hotel was full !!!

The manger was out of his wits to solve the problem. Luckily,a mathematician friend of his was in his office, and heproposed a solution the manager.

Solution:Move occupant of Room1 to Room2, move occupant ofRoom2 to Room4, and in general

Inder K. Rana, IIT Bombay – p. 16/30

December16, 2003

Hotel Hilbert

• Yet another problem:Infinite number of new visitors had arrived asking foraccommodation, and the hotel was full !!!

The manger was out of his wits to solve the problem. Luckily,a mathematician friend of his was in his office, and heproposed a solution the manager.

Solution:Move occupant of Room1 to Room2, move occupant ofRoom2 to Room4, and in general

move occupant of RoomN to Room2N.

Inder K. Rana, IIT Bombay – p. 16/30

December16, 2003

Hotel Hilbert

• Yet another problem:Infinite number of new visitors had arrived asking foraccommodation, and the hotel was full !!!

The manger was out of his wits to solve the problem. Luckily,a mathematician friend of his was in his office, and heproposed a solution the manager.

Solution:Move occupant of Room1 to Room2, move occupant ofRoom2 to Room4, and in general

move occupant of RoomN to Room2N.

This will shift all the present occupants of the hotel on theright-hand side of the hotel, i.e., in the even-numbered rooms,making all the odd numbered rooms vacant.

Inder K. Rana, IIT Bombay – p. 16/30

December16, 2003

Hotel Hilbert

• Yet another problem:Infinite number of new visitors had arrived asking foraccommodation, and the hotel was full !!!

The manger was out of his wits to solve the problem. Luckily,a mathematician friend of his was in his office, and heproposed a solution the manager.

Solution:Move occupant of Room1 to Room2, move occupant ofRoom2 to Room4, and in general

move occupant of RoomN to Room2N.

This will shift all the present occupants of the hotel on theright-hand side of the hotel, i.e., in the even-numbered rooms,making all the odd numbered rooms vacant.

Accommodate all the new guests (infinite of them) into all theodd-numbered (vacant) rooms.

Inder K. Rana, IIT Bombay – p. 16/30

December16, 2003

Hotel Hilbert

Though happy for the time being, the problems for the hotelmanager did not end there. Another problem arose whosesolution seemed impossible.

Inder K. Rana, IIT Bombay – p. 17/30

December16, 2003

Hotel Hilbert

Though happy for the time being, the problems for the hotelmanager did not end there. Another problem arose whosesolution seemed impossible.

• Super problem:Accommodation had to be arranged for the participants ofYouth Festival teams of infinite number of colleges:C1, C2, C3, . . . , each college team consisting of infinitenumber of participants, and the hotel was full !!!!

Inder K. Rana, IIT Bombay – p. 17/30

December16, 2003

Hotel Hilbert

Though happy for the time being, the problems for the hotelmanager did not end there. Another problem arose whosesolution seemed impossible.

• Super problem:Accommodation had to be arranged for the participants ofYouth Festival teams of infinite number of colleges:C1, C2, C3, . . . , each college team consisting of infinitenumber of participants, and the hotel was full !!!!

Hotel manager was fully shaken by this crises. He rushed tohis mathematician friend for help. This time themathematician friend had to scratched his head for sometime, but he did help his friend.

Inder K. Rana, IIT Bombay – p. 17/30

December16, 2003

Hotel Hilbert

Though happy for the time being, the problems for the hotelmanager did not end there. Another problem arose whosesolution seemed impossible.

• Super problem:Accommodation had to be arranged for the participants ofYouth Festival teams of infinite number of colleges:C1, C2, C3, . . . , each college team consisting of infinitenumber of participants, and the hotel was full !!!!

Hotel manager was fully shaken by this crises. He rushed tohis mathematician friend for help. This time themathematician friend had to scratched his head for sometime, but he did help his friend.

• The solution is based on the fact:there are infinite number of primes,

Inder K. Rana, IIT Bombay – p. 17/30

December16, 2003

Hotel Hilbert

Though happy for the time being, the problems for the hotelmanager did not end there. Another problem arose whosesolution seemed impossible.

• Super problem:Accommodation had to be arranged for the participants ofYouth Festival teams of infinite number of colleges:C1, C2, C3, . . . , each college team consisting of infinitenumber of participants, and the hotel was full !!!!

Hotel manager was fully shaken by this crises. He rushed tohis mathematician friend for help. This time themathematician friend had to scratched his head for sometime, but he did help his friend.

• The solution is based on the fact:there are infinite number of primes,

and the solution is as follows:

Inder K. Rana, IIT Bombay – p. 17/30

December16, 2003

Countably-infinite

• Solution:Let the primes be denoted byp0 = 2, p1 = 3, p2 = 5, p3 = 7, . . . .

Inder K. Rana, IIT Bombay – p. 18/30

December16, 2003

Countably-infinite

• Solution:Let the primes be denoted byp0 = 2, p1 = 3, p2 = 5, p3 = 7, . . . .

First step of the solution is to shift the current occupants ofRoomN to Room2N .

Inder K. Rana, IIT Bombay – p. 18/30

December16, 2003

Countably-infinite

• Solution:Let the primes be denoted byp0 = 2, p1 = 3, p2 = 5, p3 = 7, . . . .

First step of the solution is to shift the current occupants ofRoomN to Room2N .

• Note that the current occupants will get shifted to the evennumbered rooms, namely Rooms2, 4, 8, 16, 32, 64, . . . , leavingall other rooms vacant.

Inder K. Rana, IIT Bombay – p. 18/30

December16, 2003

Countably-infinite

• Solution:Let the primes be denoted byp0 = 2, p1 = 3, p2 = 5, p3 = 7, . . . .

First step of the solution is to shift the current occupants ofRoomN to Room2N .

• Note that the current occupants will get shifted to the evennumbered rooms, namely Rooms2, 4, 8, 16, 32, 64, . . . , leavingall other rooms vacant.

Next, put the jth member of the Cith team in room(pi)

j , j = i, 2, 3, . . . .

Inder K. Rana, IIT Bombay – p. 18/30

December16, 2003

Countably-infinite

• Solution:Let the primes be denoted byp0 = 2, p1 = 3, p2 = 5, p3 = 7, . . . .

First step of the solution is to shift the current occupants ofRoomN to Room2N .

• Note that the current occupants will get shifted to the evennumbered rooms, namely Rooms2, 4, 8, 16, 32, 64, . . . , leavingall other rooms vacant.

Next, put the jth member of the Cith team in room(pi)

j , j = i, 2, 3, . . . .

• Note that (pi)j is an odd number for every j.

Inder K. Rana, IIT Bombay – p. 18/30

December16, 2003

Countably-infinite

• Solution:Let the primes be denoted byp0 = 2, p1 = 3, p2 = 5, p3 = 7, . . . .

First step of the solution is to shift the current occupants ofRoomN to Room2N .

• Note that the current occupants will get shifted to the evennumbered rooms, namely Rooms2, 4, 8, 16, 32, 64, . . . , leavingall other rooms vacant.

Next, put the jth member of the Cith team in room(pi)

j , j = i, 2, 3, . . . .

• Note that (pi)j is an odd number for every j.

This solution not only accommodated all the new visitorswithout any clashes, in fact

Inder K. Rana, IIT Bombay – p. 18/30

December16, 2003

Countably-infinite

• Solution:Let the primes be denoted byp0 = 2, p1 = 3, p2 = 5, p3 = 7, . . . .

First step of the solution is to shift the current occupants ofRoomN to Room2N .

• Note that the current occupants will get shifted to the evennumbered rooms, namely Rooms2, 4, 8, 16, 32, 64, . . . , leavingall other rooms vacant.

Next, put the jth member of the Cith team in room(pi)

j , j = i, 2, 3, . . . .

• Note that (pi)j is an odd number for every j.

This solution not only accommodated all the new visitorswithout any clashes, in fact

infinite number of rooms, for example Rooms6, 12, 18, . . . orRooms10, 20, 30, . . . , remained vacant.

Inder K. Rana, IIT Bombay – p. 18/30

December16, 2003

Countably-infinite

• Thus, even though initially no room was vacant, hotel was ableto accommodate infinite number of teams, each having infinitemembers, and still leaving infinite number of rooms vacant!

Inder K. Rana, IIT Bombay – p. 19/30

December16, 2003

Countably-infinite

• Thus, even though initially no room was vacant, hotel was ableto accommodate infinite number of teams, each having infinitemembers, and still leaving infinite number of rooms vacant!

The solutions of the above problems are in fact theorems inset theory. Let us first make the following:

Inder K. Rana, IIT Bombay – p. 19/30

December16, 2003

Countably-infinite

• Thus, even though initially no room was vacant, hotel was ableto accommodate infinite number of teams, each having infinitemembers, and still leaving infinite number of rooms vacant!

The solutions of the above problems are in fact theorems inset theory. Let us first make the following:

• Definition:

Inder K. Rana, IIT Bombay – p. 19/30

December16, 2003

Countably-infinite

• Thus, even though initially no room was vacant, hotel was ableto accommodate infinite number of teams, each having infinitemembers, and still leaving infinite number of rooms vacant!

The solutions of the above problems are in fact theorems inset theory. Let us first make the following:

• Definition:

1. A nonempty set B is said to be countably infinite if thereexists a bijective map ψ : B → IN.

Inder K. Rana, IIT Bombay – p. 19/30

December16, 2003

Countably-infinite

• Thus, even though initially no room was vacant, hotel was ableto accommodate infinite number of teams, each having infinitemembers, and still leaving infinite number of rooms vacant!

The solutions of the above problems are in fact theorems inset theory. Let us first make the following:

• Definition:

1. A nonempty set B is said to be countably infinite if thereexists a bijective map ψ : B → IN.

2. A set C is said to be countable if either it is countably finiteor countably infinite.

Inder K. Rana, IIT Bombay – p. 19/30

December16, 2003

Countably-infinite

• Thus, even though initially no room was vacant, hotel was ableto accommodate infinite number of teams, each having infinitemembers, and still leaving infinite number of rooms vacant!

The solutions of the above problems are in fact theorems inset theory. Let us first make the following:

• Definition:

1. A nonempty set B is said to be countably infinite if thereexists a bijective map ψ : B → IN.

2. A set C is said to be countable if either it is countably finiteor countably infinite.

• Theorem:

Inder K. Rana, IIT Bombay – p. 19/30

December16, 2003

Countably-infinite

• Thus, even though initially no room was vacant, hotel was ableto accommodate infinite number of teams, each having infinitemembers, and still leaving infinite number of rooms vacant!

The solutions of the above problems are in fact theorems inset theory. Let us first make the following:

• Definition:

1. A nonempty set B is said to be countably infinite if thereexists a bijective map ψ : B → IN.

2. A set C is said to be countable if either it is countably finiteor countably infinite.

• Theorem:

(i) If A is countably finite and B is countably infinite, A ∪B is countablyinfinite.

Inder K. Rana, IIT Bombay – p. 19/30

December16, 2003

Countably-infinite

• Thus, even though initially no room was vacant, hotel was ableto accommodate infinite number of teams, each having infinitemembers, and still leaving infinite number of rooms vacant!

The solutions of the above problems are in fact theorems inset theory. Let us first make the following:

• Definition:

1. A nonempty set B is said to be countably infinite if thereexists a bijective map ψ : B → IN.

2. A set C is said to be countable if either it is countably finiteor countably infinite.

• Theorem:

(i) If A is countably finite and B is countably infinite, A ∪B is countablyinfinite.

(ii) If A and B are countably infinite, so also is A ∪B.

Inder K. Rana, IIT Bombay – p. 19/30

December16, 2003

Countably-infinite

(iii) If An is a countable set for each n ∈ IN, then A :=⋃

n=1An is also acountable set.

Inder K. Rana, IIT Bombay – p. 20/30

December16, 2003

Countably-infinite

(iii) If An is a countable set for each n ∈ IN, then A :=⋃

n=1An is also acountable set.

• Examples:

Inder K. Rana, IIT Bombay – p. 20/30

December16, 2003

Countably-infinite

(iii) If An is a countable set for each n ∈ IN, then A :=⋃

n=1An is also acountable set.

• Examples:

1. The set IN itself is countably infinite.

Inder K. Rana, IIT Bombay – p. 20/30

December16, 2003

Countably-infinite

(iii) If An is a countable set for each n ∈ IN, then A :=⋃

n=1An is also acountable set.

• Examples:

1. The set IN itself is countably infinite.2. INe, the set of all even natural numbers and INo, the set of

all odd natural numbers are both countably infinite.

Inder K. Rana, IIT Bombay – p. 20/30

December16, 2003

Countably-infinite

(iii) If An is a countable set for each n ∈ IN, then A :=⋃

n=1An is also acountable set.

• Examples:

1. The set IN itself is countably infinite.2. INe, the set of all even natural numbers and INo, the set of

all odd natural numbers are both countably infinite.3. The set IN × IN is countably infinite since g : IN × IN → IN

defined byg(m,n) := 2m−1(2n− 1),

is a bijective map.

Inder K. Rana, IIT Bombay – p. 20/30

December16, 2003

Countably-infinite

(iii) If An is a countable set for each n ∈ IN, then A :=⋃

n=1An is also acountable set.

• Examples:

1. The set IN itself is countably infinite.2. INe, the set of all even natural numbers and INo, the set of

all odd natural numbers are both countably infinite.3. The set IN × IN is countably infinite since g : IN × IN → IN

defined byg(m,n) := 2m−1(2n− 1),

is a bijective map.4. The sets ZZ and IQ are countably infinite.

Inder K. Rana, IIT Bombay – p. 20/30

December16, 2003

Countably-infinite

(iii) If An is a countable set for each n ∈ IN, then A :=⋃

n=1An is also acountable set.

• Examples:

1. The set IN itself is countably infinite.2. INe, the set of all even natural numbers and INo, the set of

all odd natural numbers are both countably infinite.3. The set IN × IN is countably infinite since g : IN × IN → IN

defined byg(m,n) := 2m−1(2n− 1),

is a bijective map.4. The sets ZZ and IQ are countably infinite.5. The set of all algebraic numbers is countably infinite.

Inder K. Rana, IIT Bombay – p. 20/30

December16, 2003

Countably-infinite

(iii) If An is a countable set for each n ∈ IN, then A :=⋃

n=1An is also acountable set.

• Examples:

1. The set IN itself is countably infinite.2. INe, the set of all even natural numbers and INo, the set of

all odd natural numbers are both countably infinite.3. The set IN × IN is countably infinite since g : IN × IN → IN

defined byg(m,n) := 2m−1(2n− 1),

is a bijective map.4. The sets ZZ and IQ are countably infinite.5. The set of all algebraic numbers is countably infinite.

Inder K. Rana, IIT Bombay – p. 20/30

December16, 2003

Cantor’s Theorem

• Question:Is P(IN), the set of all subsets of IN, countably infinite?

Inder K. Rana, IIT Bombay – p. 21/30

December16, 2003

Cantor’s Theorem

• Question:Is P(IN), the set of all subsets of IN, countably infinite?

To answer this we need the following:

Inder K. Rana, IIT Bombay – p. 21/30

December16, 2003

Cantor’s Theorem

• Question:Is P(IN), the set of all subsets of IN, countably infinite?

To answer this we need the following:• Theorem (Cantor):

Let X be any nonempty set. Then there exists a one-one mapping of Xinto P(X) but there exists no bijection between X and P(X).

Inder K. Rana, IIT Bombay – p. 21/30

December16, 2003

Cantor’s Theorem

• Question:Is P(IN), the set of all subsets of IN, countably infinite?

To answer this we need the following:• Theorem (Cantor):

Let X be any nonempty set. Then there exists a one-one mapping of Xinto P(X) but there exists no bijection between X and P(X).

• Click here to see the proof.

Inder K. Rana, IIT Bombay – p. 21/30

December16, 2003

Cantor’s Theorem

• Question:Is P(IN), the set of all subsets of IN, countably infinite?

To answer this we need the following:• Theorem (Cantor):

Let X be any nonempty set. Then there exists a one-one mapping of Xinto P(X) but there exists no bijection between X and P(X).

• Click here to see the proof.

As a consequence of this, we have:

Inder K. Rana, IIT Bombay – p. 21/30

December16, 2003

Cantor’s Theorem

• Question:Is P(IN), the set of all subsets of IN, countably infinite?

To answer this we need the following:• Theorem (Cantor):

Let X be any nonempty set. Then there exists a one-one mapping of Xinto P(X) but there exists no bijection between X and P(X).

• Click here to see the proof.

As a consequence of this, we have:• Corollary:

If X is a countably infinite set, then P(X) is not countably infinite.

Inder K. Rana, IIT Bombay – p. 21/30

December16, 2003

Cantor’s Theorem

• Question:Is P(IN), the set of all subsets of IN, countably infinite?

To answer this we need the following:• Theorem (Cantor):

Let X be any nonempty set. Then there exists a one-one mapping of Xinto P(X) but there exists no bijection between X and P(X).

• Click here to see the proof.

As a consequence of this, we have:• Corollary:

If X is a countably infinite set, then P(X) is not countably infinite.

• Thus, in some sense P(IN) has much more number ofelements than that of IN.

Inder K. Rana, IIT Bombay – p. 21/30

December16, 2003

Cantor’s Theorem

• Question:Is P(IN), the set of all subsets of IN, countably infinite?

To answer this we need the following:• Theorem (Cantor):

Let X be any nonempty set. Then there exists a one-one mapping of Xinto P(X) but there exists no bijection between X and P(X).

• Click here to see the proof.

As a consequence of this, we have:• Corollary:

If X is a countably infinite set, then P(X) is not countably infinite.

• Thus, in some sense P(IN) has much more number ofelements than that of IN.

Let us give a name to the ’number’ of elements of a set.

Inder K. Rana, IIT Bombay – p. 21/30

December16, 2003

Cardinality

• Definition:Let X be any nonempty set.

Inder K. Rana, IIT Bombay – p. 22/30

December16, 2003

Cardinality

• Definition:Let X be any nonempty set.

(i) If X is a finite set, we define its cardinality to be the numberof elements in X,and is denoted it by that natural number.

Inder K. Rana, IIT Bombay – p. 22/30

December16, 2003

Cardinality

• Definition:Let X be any nonempty set.

(i) If X is a finite set, we define its cardinality to be the numberof elements in X,and is denoted it by that natural number.

(ii) If X is countably infinite, we say X has cardinalityaleph-nought and denote it by ℵ0.

Inder K. Rana, IIT Bombay – p. 22/30

December16, 2003

Cardinality

• Definition:Let X be any nonempty set.

(i) If X is a finite set, we define its cardinality to be the numberof elements in X,and is denoted it by that natural number.

(ii) If X is countably infinite, we say X has cardinalityaleph-nought and denote it by ℵ0.

The question arises:What is the cardinality of a set which is not countable?

Inder K. Rana, IIT Bombay – p. 22/30

December16, 2003

Cardinality

• Definition:Let X be any nonempty set.

(i) If X is a finite set, we define its cardinality to be the numberof elements in X,and is denoted it by that natural number.

(ii) If X is countably infinite, we say X has cardinalityaleph-nought and denote it by ℵ0.

The question arises:What is the cardinality of a set which is not countable?

To answer this, let us note that if X is a finite set with melements, then P(X) is also finite with 2m elements.Thus wemake the following:

Inder K. Rana, IIT Bombay – p. 22/30

December16, 2003

Cardinality

• Definition:Let X be any nonempty set.

(i) If X is a finite set, we define its cardinality to be the numberof elements in X,and is denoted it by that natural number.

(ii) If X is countably infinite, we say X has cardinalityaleph-nought and denote it by ℵ0.

The question arises:What is the cardinality of a set which is not countable?

To answer this, let us note that if X is a finite set with melements, then P(X) is also finite with 2m elements.Thus wemake the following:

• Definition:For a set X whose cardinality is defined and is card(X), let usdenote the cardinality of P(X) by 2card(X).

Inder K. Rana, IIT Bombay – p. 22/30

December16, 2003

Infinity of Infinities

Thus,

card(IN) = ℵ0, card(P(IN)) = 2ℵ0 , card(P(P(IN))) = 22ℵ

0 ,

and so on.

Inder K. Rana, IIT Bombay – p. 23/30

December16, 2003

Infinity of Infinities

Thus,

card(IN) = ℵ0, card(P(IN)) = 2ℵ0 , card(P(P(IN))) = 22ℵ

0 ,

and so on.

Note that, ℵ0, 2ℵ0 , 22ℵ

0 , . . . , are all infinities.

Inder K. Rana, IIT Bombay – p. 23/30

December16, 2003

Infinity of Infinities

Thus,

card(IN) = ℵ0, card(P(IN)) = 2ℵ0 , card(P(P(IN))) = 22ℵ

0 ,

and so on.

Note that, ℵ0, 2ℵ0 , 22ℵ

0 , . . . , are all infinities.• Thus, we have infinity of infinities: with

1 < 2 < 3 < · · · < ℵ0 < 2ℵ0 := c < 2c < 22c

< · · · .

Inder K. Rana, IIT Bombay – p. 23/30

December16, 2003

Infinity of Infinities

Thus,

card(IN) = ℵ0, card(P(IN)) = 2ℵ0 , card(P(P(IN))) = 22ℵ

0 ,

and so on.

Note that, ℵ0, 2ℵ0 , 22ℵ

0 , . . . , are all infinities.• Thus, we have infinity of infinities: with

1 < 2 < 3 < · · · < ℵ0 < 2ℵ0 := c < 2c < 22c

< · · · .

• Thus, there is counting beyond infinity.

Inder K. Rana, IIT Bombay – p. 23/30

December16, 2003

Infinity of Infinities

Thus,

card(IN) = ℵ0, card(P(IN)) = 2ℵ0 , card(P(P(IN))) = 22ℵ

0 ,

and so on.

Note that, ℵ0, 2ℵ0 , 22ℵ

0 , . . . , are all infinities.• Thus, we have infinity of infinities: with

1 < 2 < 3 < · · · < ℵ0 < 2ℵ0 := c < 2c < 22c

< · · · .

• Thus, there is counting beyond infinity.• The above process of counting enables us to define the

cardinality (the number of elements ) for a set which is eitherfinite, or countably-infinite, or for the power set of a set, forwhich the notion of cardinality has already been defined.

Inder K. Rana, IIT Bombay – p. 23/30

December16, 2003

Infinity of Infinities

Thus,

card(IN) = ℵ0, card(P(IN)) = 2ℵ0 , card(P(P(IN))) = 22ℵ

0 ,

and so on.

Note that, ℵ0, 2ℵ0 , 22ℵ

0 , . . . , are all infinities.• Thus, we have infinity of infinities: with

1 < 2 < 3 < · · · < ℵ0 < 2ℵ0 := c < 2c < 22c

< · · · .

• Thus, there is counting beyond infinity.• The above process of counting enables us to define the

cardinality (the number of elements ) for a set which is eitherfinite, or countably-infinite, or for the power set of a set, forwhich the notion of cardinality has already been defined.

Inder K. Rana, IIT Bombay – p. 23/30

December16, 2003

Equivalance of sets

• This raises the following natural questions:

Inder K. Rana, IIT Bombay – p. 24/30

December16, 2003

Equivalance of sets

• This raises the following natural questions:• How do we compare the number of elements for given sets?

Inder K. Rana, IIT Bombay – p. 24/30

December16, 2003

Equivalance of sets

• This raises the following natural questions:• How do we compare the number of elements for given sets?• How to define the cardinality of any given set which is not of

the type as described above?

Inder K. Rana, IIT Bombay – p. 24/30

December16, 2003

Equivalance of sets

• This raises the following natural questions:• How do we compare the number of elements for given sets?• How to define the cardinality of any given set which is not of

the type as described above?

Assuming that one already has the notion of cardinalitydefined for every set, it is easy to answer the secondquestion.

Inder K. Rana, IIT Bombay – p. 24/30

December16, 2003

Equivalance of sets

• This raises the following natural questions:• How do we compare the number of elements for given sets?• How to define the cardinality of any given set which is not of

the type as described above?

Assuming that one already has the notion of cardinalitydefined for every set, it is easy to answer the secondquestion.

Definition:Given two sets A and B, we say A and B have the samecardinality if there exists a one-one, onto map f : A→ B. Wewrite this as A ∼ B.

Inder K. Rana, IIT Bombay – p. 24/30

December16, 2003

Equivalance of sets

• This raises the following natural questions:• How do we compare the number of elements for given sets?• How to define the cardinality of any given set which is not of

the type as described above?

Assuming that one already has the notion of cardinalitydefined for every set, it is easy to answer the secondquestion.

Definition:Given two sets A and B, we say A and B have the samecardinality if there exists a one-one, onto map f : A→ B. Wewrite this as A ∼ B.

• On any given collection S of sets, the relation ∼ is anequivalence relation.

Inder K. Rana, IIT Bombay – p. 24/30

December16, 2003

CBS-theorem

• Examples:

Inder K. Rana, IIT Bombay – p. 25/30

December16, 2003

CBS-theorem

• Examples:

1. For n ∈ IN, {1, 2, . . . , n} ∼ {1, 2, . . . ,m} iff n = m.

Inder K. Rana, IIT Bombay – p. 25/30

December16, 2003

CBS-theorem

• Examples:

1. For n ∈ IN, {1, 2, . . . , n} ∼ {1, 2, . . . ,m} iff n = m.

2. The set IN is not equivalent to [1, n] for any n ∈ IN.

Inder K. Rana, IIT Bombay – p. 25/30

December16, 2003

CBS-theorem

• Examples:

1. For n ∈ IN, {1, 2, . . . , n} ∼ {1, 2, . . . ,m} iff n = m.

2. The set IN is not equivalent to [1, n] for any n ∈ IN.

3. Clearly any two of the sets IN, ZZ+, ZZ−, IQ and IN × IN areequivalent to one another.

Inder K. Rana, IIT Bombay – p. 25/30

December16, 2003

CBS-theorem

• Examples:

1. For n ∈ IN, {1, 2, . . . , n} ∼ {1, 2, . . . ,m} iff n = m.

2. The set IN is not equivalent to [1, n] for any n ∈ IN.

3. Clearly any two of the sets IN, ZZ+, ZZ−, IQ and IN × IN areequivalent to one another.

4. For a ∈ IR, a > 0, (0, 1) ∼ (0, a).

Inder K. Rana, IIT Bombay – p. 25/30

December16, 2003

CBS-theorem

• Examples:

1. For n ∈ IN, {1, 2, . . . , n} ∼ {1, 2, . . . ,m} iff n = m.

2. The set IN is not equivalent to [1, n] for any n ∈ IN.

3. Clearly any two of the sets IN, ZZ+, ZZ−, IQ and IN × IN areequivalent to one another.

4. For a ∈ IR, a > 0, (0, 1) ∼ (0, a).

5. (0, 1) ∼ IR, since x 7−→x

|x| + 1from IR to (0, 1) is a one-one,

onto map.

Inder K. Rana, IIT Bombay – p. 25/30

December16, 2003

CBS-theorem

• Examples:

1. For n ∈ IN, {1, 2, . . . , n} ∼ {1, 2, . . . ,m} iff n = m.

2. The set IN is not equivalent to [1, n] for any n ∈ IN.

3. Clearly any two of the sets IN, ZZ+, ZZ−, IQ and IN × IN areequivalent to one another.

4. For a ∈ IR, a > 0, (0, 1) ∼ (0, a).

5. (0, 1) ∼ IR, since x 7−→x

|x| + 1from IR to (0, 1) is a one-one,

onto map.

• An important theorem which allows one to check theequivalence of two sets is the following:Theorem (Cantor-Bernstein-Schr oder):If X and Y are two nonempty sets such that each is equivalent to a subsetof the other then they are equivalent.

Inder K. Rana, IIT Bombay – p. 25/30

December16, 2003

CBS-theorem

• Examples:

1. For n ∈ IN, {1, 2, . . . , n} ∼ {1, 2, . . . ,m} iff n = m.

2. The set IN is not equivalent to [1, n] for any n ∈ IN.

3. Clearly any two of the sets IN, ZZ+, ZZ−, IQ and IN × IN areequivalent to one another.

4. For a ∈ IR, a > 0, (0, 1) ∼ (0, a).

5. (0, 1) ∼ IR, since x 7−→x

|x| + 1from IR to (0, 1) is a one-one,

onto map.

• An important theorem which allows one to check theequivalence of two sets is the following:Theorem (Cantor-Bernstein-Schr oder):If X and Y are two nonempty sets such that each is equivalent to a subsetof the other then they are equivalent.

• Click here to see the proof.

Inder K. Rana, IIT Bombay – p. 25/30

December16, 2003

2ℵ0 = c

Examples:

Inder K. Rana, IIT Bombay – p. 26/30

December16, 2003

2ℵ0 = c

Examples:

1. Let X be any subset of IR such that X includes a nonemptyopen interval. Then X ∼ IR.

Inder K. Rana, IIT Bombay – p. 26/30

December16, 2003

2ℵ0 = c

Examples:

1. Let X be any subset of IR such that X includes a nonemptyopen interval. Then X ∼ IR.

To see this, let (a, b) ⊆ X. Since X ∼ X ⊆ IR andIR ∼ (0, 1) ∼⊆ X, by CBS- theorem, X ∼ IR.

Inder K. Rana, IIT Bombay – p. 26/30

December16, 2003

2ℵ0 = c

Examples:

1. Let X be any subset of IR such that X includes a nonemptyopen interval. Then X ∼ IR.

To see this, let (a, b) ⊆ X. Since X ∼ X ⊆ IR andIR ∼ (0, 1) ∼⊆ X, by CBS- theorem, X ∼ IR.

2. Any two intervals in IR are equivalent.

Inder K. Rana, IIT Bombay – p. 26/30

December16, 2003

2ℵ0 = c

Examples:

1. Let X be any subset of IR such that X includes a nonemptyopen interval. Then X ∼ IR.

To see this, let (a, b) ⊆ X. Since X ∼ X ⊆ IR andIR ∼ (0, 1) ∼⊆ X, by CBS- theorem, X ∼ IR.

2. Any two intervals in IR are equivalent.3. P(IN) ∼ I := (0, 1) ∼ IR.

Inder K. Rana, IIT Bombay – p. 26/30

December16, 2003

2ℵ0 = c

Examples:

1. Let X be any subset of IR such that X includes a nonemptyopen interval. Then X ∼ IR.

To see this, let (a, b) ⊆ X. Since X ∼ X ⊆ IR andIR ∼ (0, 1) ∼⊆ X, by CBS- theorem, X ∼ IR.

2. Any two intervals in IR are equivalent.3. P(IN) ∼ I := (0, 1) ∼ IR.

To see this let A ∈ P(IN) and let p(A) ∈ I be the point withdecimal expansion p(A) = ·x1x2x3 . . . where xi = 1 if i ∈ Aand xi = 0 if i 6∈ A.

Inder K. Rana, IIT Bombay – p. 26/30

December16, 2003

2ℵ0 = c

Examples:

1. Let X be any subset of IR such that X includes a nonemptyopen interval. Then X ∼ IR.

To see this, let (a, b) ⊆ X. Since X ∼ X ⊆ IR andIR ∼ (0, 1) ∼⊆ X, by CBS- theorem, X ∼ IR.

2. Any two intervals in IR are equivalent.3. P(IN) ∼ I := (0, 1) ∼ IR.

To see this let A ∈ P(IN) and let p(A) ∈ I be the point withdecimal expansion p(A) = ·x1x2x3 . . . where xi = 1 if i ∈ Aand xi = 0 if i 6∈ A.

Then p is a one-one function from P(IN) to I.

Inder K. Rana, IIT Bombay – p. 26/30

December16, 2003

2ℵ0 = c

Examples:

1. Let X be any subset of IR such that X includes a nonemptyopen interval. Then X ∼ IR.

To see this, let (a, b) ⊆ X. Since X ∼ X ⊆ IR andIR ∼ (0, 1) ∼⊆ X, by CBS- theorem, X ∼ IR.

2. Any two intervals in IR are equivalent.3. P(IN) ∼ I := (0, 1) ∼ IR.

To see this let A ∈ P(IN) and let p(A) ∈ I be the point withdecimal expansion p(A) = ·x1x2x3 . . . where xi = 1 if i ∈ Aand xi = 0 if i 6∈ A.

Then p is a one-one function from P(IN) to I.Also for any x ∈ (0, 1), let x = ·a1a2a3 . . . be its decimalexpansion. Let s(x) := {n ∈ IN|an = 1}.

Inder K. Rana, IIT Bombay – p. 26/30

December16, 2003

2ℵ0 = c

Examples:

1. Let X be any subset of IR such that X includes a nonemptyopen interval. Then X ∼ IR.

To see this, let (a, b) ⊆ X. Since X ∼ X ⊆ IR andIR ∼ (0, 1) ∼⊆ X, by CBS- theorem, X ∼ IR.

2. Any two intervals in IR are equivalent.3. P(IN) ∼ I := (0, 1) ∼ IR.

To see this let A ∈ P(IN) and let p(A) ∈ I be the point withdecimal expansion p(A) = ·x1x2x3 . . . where xi = 1 if i ∈ Aand xi = 0 if i 6∈ A.

Then p is a one-one function from P(IN) to I.Also for any x ∈ (0, 1), let x = ·a1a2a3 . . . be its decimalexpansion. Let s(x) := {n ∈ IN|an = 1}.

Then s(x) ∈ P(IN) and clearly the map x 7−→ s(x) is aone-one function from I into P(X).

Inder K. Rana, IIT Bombay – p. 26/30

December16, 2003

2ℵ0 = c

Examples:

1. Let X be any subset of IR such that X includes a nonemptyopen interval. Then X ∼ IR.

To see this, let (a, b) ⊆ X. Since X ∼ X ⊆ IR andIR ∼ (0, 1) ∼⊆ X, by CBS- theorem, X ∼ IR.

2. Any two intervals in IR are equivalent.3. P(IN) ∼ I := (0, 1) ∼ IR.

To see this let A ∈ P(IN) and let p(A) ∈ I be the point withdecimal expansion p(A) = ·x1x2x3 . . . where xi = 1 if i ∈ Aand xi = 0 if i 6∈ A.

Then p is a one-one function from P(IN) to I.Also for any x ∈ (0, 1), let x = ·a1a2a3 . . . be its decimalexpansion. Let s(x) := {n ∈ IN|an = 1}.

Then s(x) ∈ P(IN) and clearly the map x 7−→ s(x) is aone-one function from I into P(X).

Hence, by the CBS-theorem I ∼ P(IN).Inder K. Rana, IIT Bombay – p. 26/30

December16, 2003

Dedekind-infinite

• The notion of ’cardinality of a set’ or of ’cardinal number’requires more elaborate discussion, see

"Naive Set Theory by P. R. Halmos.

Inder K. Rana, IIT Bombay – p. 27/30

December16, 2003

Dedekind-infinite

• The notion of ’cardinality of a set’ or of ’cardinal number’requires more elaborate discussion, see

"Naive Set Theory by P. R. Halmos.

• In our discussions, defined countably-finite andcountably-infinite in comparison with the set of naturalnumbers.

Inder K. Rana, IIT Bombay – p. 27/30

December16, 2003

Dedekind-infinite

• The notion of ’cardinality of a set’ or of ’cardinal number’requires more elaborate discussion, see

"Naive Set Theory by P. R. Halmos.

• In our discussions, defined countably-finite andcountably-infinite in comparison with the set of naturalnumbers.

An ’intrinsic definition’ of a set to be infinite, which does notmake use of natural numbers, was given by RichardDedekind:

Inder K. Rana, IIT Bombay – p. 27/30

December16, 2003

Dedekind-infinite

• The notion of ’cardinality of a set’ or of ’cardinal number’requires more elaborate discussion, see

"Naive Set Theory by P. R. Halmos.

• In our discussions, defined countably-finite andcountably-infinite in comparison with the set of naturalnumbers.

An ’intrinsic definition’ of a set to be infinite, which does notmake use of natural numbers, was given by RichardDedekind:

• Definition:A set is called an infinite set if it can be put in one-to-onecorrespondence with a proper subset of itself. And a set iscalled finite if it is not infinite.

Inder K. Rana, IIT Bombay – p. 27/30

December16, 2003

Dedekind-infinite

• Examples:

Inder K. Rana, IIT Bombay – p. 28/30

December16, 2003

Dedekind-infinite

• Examples:

1. IN is infinite as it is in one-to-one correspondence with theproper subset INo.

Inder K. Rana, IIT Bombay – p. 28/30

December16, 2003

Dedekind-infinite

• Examples:

1. IN is infinite as it is in one-to-one correspondence with theproper subset INo.

2. Each of , ZZ , IQ, and IR is an infinite set.

Inder K. Rana, IIT Bombay – p. 28/30

December16, 2003

Dedekind-infinite

• Examples:

1. IN is infinite as it is in one-to-one correspondence with theproper subset INo.

2. Each of , ZZ , IQ, and IR is an infinite set.3. For every n ∈ IN, the set {1, 2, . . . , n} is finite.

Inder K. Rana, IIT Bombay – p. 28/30

December16, 2003

Dedekind-Infinite

• "The Whole is there, The Whole is here,

Inder K. Rana, IIT Bombay – p. 29/30

December16, 2003

Dedekind-Infinite

• "The Whole is there, The Whole is here,

From the Whole emanates the Whole,

Inder K. Rana, IIT Bombay – p. 29/30

December16, 2003

Dedekind-Infinite

• "The Whole is there, The Whole is here,

From the Whole emanates the Whole,

Taking away the Whole from the Whole,

Inder K. Rana, IIT Bombay – p. 29/30

December16, 2003

Dedekind-Infinite

• "The Whole is there, The Whole is here,

From the Whole emanates the Whole,

Taking away the Whole from the Whole,

What remains is the Whole."

Inder K. Rana, IIT Bombay – p. 29/30

December16, 2003

Set Theory

Thus we have two ways of saying a set is finite or not.

Inder K. Rana, IIT Bombay – p. 30/30

December16, 2003

Set Theory

Thus we have two ways of saying a set is finite or not.

The natural question arises:

Are these two definitions equivalent?

Inder K. Rana, IIT Bombay – p. 30/30

December16, 2003

Set Theory

Thus we have two ways of saying a set is finite or not.

The natural question arises:

Are these two definitions equivalent?• Answer to this question leads to some fundamental concepts

and questions in set theory, like:

Inder K. Rana, IIT Bombay – p. 30/30

December16, 2003

Set Theory

Thus we have two ways of saying a set is finite or not.

The natural question arises:

Are these two definitions equivalent?• Answer to this question leads to some fundamental concepts

and questions in set theory, like:

What is a set?

Inder K. Rana, IIT Bombay – p. 30/30

December16, 2003

Set Theory

Thus we have two ways of saying a set is finite or not.

The natural question arises:

Are these two definitions equivalent?• Answer to this question leads to some fundamental concepts

and questions in set theory, like:

What is a set?

What are natural numbers?

Inder K. Rana, IIT Bombay – p. 30/30

December16, 2003

Set Theory

Thus we have two ways of saying a set is finite or not.

The natural question arises:

Are these two definitions equivalent?• Answer to this question leads to some fundamental concepts

and questions in set theory, like:

What is a set?

What are natural numbers?

In case you would like to know more about this, have a look atthe following book:

Inder K. Rana, IIT Bombay – p. 30/30

December16, 2003

Set Theory

Thus we have two ways of saying a set is finite or not.

The natural question arises:

Are these two definitions equivalent?• Answer to this question leads to some fundamental concepts

and questions in set theory, like:

What is a set?

What are natural numbers?

In case you would like to know more about this, have a look atthe following book:

FROM NUMBERS TO ANALYSIS

INDER K. RANA

WORLD SCIENTIFIC PRESS, SINGAPORE

Inder K. Rana, IIT Bombay – p. 30/30