Discrete Structures & Algorithms

Basics of Set Theory

EECE 320 — UBC

 Set Theory: Definitions and notation

A set is an unordered collection of elements.

Some examples

• {1, 2, 3} is the set containing “1” and “2” and “3.”

• {1, 1, 2, 3, 3} = {1, 2, 3} since repetition is irrelevant.
• {1, 2, 3} = {3, 2, 1} since sets are unordered.
• {1, 2, 3, …} is a way we denote an infinite set (in this case, the natural
numbers).
  = {} is the empty set, or the set containing no elements.

Note:   {} 2
 Definitions and notation

x  S means “x is an element of set S.”

x  S means “x is not an element of set S.”

A  B means “A is a subset of B.”

or, “B contains A.”
or, “every element of A is also in B.”
or, x ((x  A)  (x  B)).

A Venn
Diagram
B

3
 Definitions and notation

A  B means “A is a subset of B.”

A  B means “A is a superset of B.”
A = B if and only if A and B have exactly the same
elements.

iff, A  B and B  A
iff, A  B and A  B
iff, x ((x  A)  (x  B)).

So to show equality of sets A and B, show:

AB
BA

4
 Definitions and notation

A  B means “A is a proper subset of B.”

– A  B, and A  B.
 x ((x  A)  (x  B))  x ((x  B)  (x  A))
 x ((x  A)  (x  B))  x ((x  B) v (x  A))
 x ((x  A)  (x  B))  x ((x  B)  (x  A))
 x ((x  A)  (x  B))  x ((x  B)  (x  A))

A
B

5
 Definitions and notation

Quick examples:
• {1,2,3}  {1,2,3,4,5}
• {1,2,3}  {1,2,3,4,5}

Is   {1,2,3}? Yes! x (x  )  (x  {1,2,3})

holds, because (x  ) is false.
Vacuously
Is   {1,2,3}? No!
Is   {,1,2,3}? Yes!

6
 Definitions and notation

Quiz Time

Is {x}  {x}? Yes

Is {x}  {x,{x}}? Yes

Is {x}  {x,{x}}? Yes

Is {x}  {x}? No

7
 How to specify sets

• Explicitly: {John, Paul, George, Ringo} : and | are read “such

that” or “where”
• Implicitly: {1,2,3,…}, or {2,3,5,7,11,13,17,…}
• Set builder: { x : x is prime }, { x | x is odd }. In general
{ x : P(x) is true }, where P(x) is some description of the
set.

Example: Let D(x,y) denote “x is divisible by y.”

Give another name for
{ x : y ((y > 1)  (y < x))  D(x,y) }.
Primes

Can we use any predicate P to define a set S =

{ x : P(x) }?
8
 Predicates for defining sets

Can we use any predicate P to define a set

S = { x : P(x) }? No!

Define S = { x : x is a set where x  x }

Then, if S  S, then by the definition of S, S  S.

So S must not be in
S, right?
But, if S  S, then by the definition of S, S  S.
Doh!
There is a town with a barber who shaves all the people (and only the people)
who do not shave themselves.
Who shaves the 9
barber?
 Cardinality of sets

If S is finite, then the cardinality of S, |S|, is the number of

distinct elements in S.

If S = {1,2,3}, |S| = 3.
If S = {3,3,3,3,3}, |S| = 1.
If S = , |S| = 0.
If S = { , {}, {,{}} }, |S| = 3.
If S = {0,1,2,3,…}, |S| is infinite. (more on this later)

10
 Power sets

If S is a set, then the power set of S is

2S = { x : x  S }. or P(S)
We say, “P(S) is
If S = {a}, 2S = {, {a}}.
the set of all
If S = {a,b}, 2S = {, {a}, {b}, {a,b}}. subsets of S.”
If S = , 2S = {}.
If S = {,{}}, 2S = {, {}, {{}}, {,
{}}}.

Fact: if S is finite, |2S| = 2|S|. (If |S| = n, |2S| = 2n.)

11
 Cartesian product

The Cartesian product of two sets A and B is:

A x B = { <a,b> : a  A  b  B}

If A = {Charlie, Lucy, Linus}, and We’ll use these

B = {Brown, VanPelt}, then special sets soon!

A x B = {<Charlie, Brown>, <Lucy, Brown>, <Linus,

Brown>, <Charlie, VanPelt>, <Lucy, VanPelt>, <Linus,
VanPelt>}

A1 x A2 x … x An = {<a1, a2,…, an>: a1  A1, a2  A2, …, an

 An} a) AxB
b) |A|+|B|
A,B finite  |AxB| = ? c) |A+B|
d) |A||B| 12
 Operators

The union of two sets A and B is:

A  B = { x : x  A v x  B}

If A = {Charlie, Lucy, Linus}, and B =

{Lucy, Desi}, then
A  B = {Charlie, Lucy, Linus, Desi}

B A

13
 Operators

The intersection of two sets A and B is:

A  B = { x : x  A  x  B}

If A = {Charlie, Lucy, Linus}, and B =

{Lucy, Desi}, then
A  B = {Lucy}

B A

14
 Operators

The intersection of two sets A and B is:

A  B = { x : x  A  x  B}

If A = {x : x is a US president}, and B =
{x : x is deceased}, then
A  B = {x : x is a deceased US president}

B A

15
 Operators

The intersection of two sets A and B is:

A  B = { x : x  A  x  B}

If A = {x : x is a US president}, and B =
{x : x is in this room}, then
A  B = {x : x is a US president in this room} = 

Sets whose
B A intersection is empty
are called disjoint
sets

16
 Operators

The complement of a set A is QuickTime™ and a

TIFF (Uncompressed) decompressor

Ac = { x : x  A}
are needed to see this picture.

If A = {x : x is bored}, then

A = {x : x is not bored} =

U
A
c = U
and
Uc = 
17
 Operators

The set difference, A - B, is:

U
A
B

A-B={x:xAxB}
A - B = A  Bc

18
 Operators

The symmetric difference, A  B, is:

A  B = { x : (x  A  x  B) v (x  B  x  A)}

like “exclusive
= (A - B) U (B - A) or”

U
A
B

19
 Operators

A  B = { x : (x  A  x  B) v (x  B  x  A)}

= (A - B) U (B - A)

Proof: { x : (x  A  x  B) v (x  B  x  A)}
= { x : (x  A - B) v (x  B - A)}
= { x : x  ((A - B) U (B - A))}
= (A - B) U (B - A)

20
 Famous identities

• Two pages of (almost) obvious equivalences.

• One page of HS algebra.

Don’t
• Some new material? memorize
them,
understand
them!

They’re in Rosen,
Sec. 2.2

21
 Identities

• Identity AU=A
AU=A

• Domination AUU=U
A=

• Idempotent
AUA=A
AA=A

22
 Identities

• Excluded middle AUA=U

• Uniqueness AA=

• Double complement

A=A

23
 Identities

• Commutativity AUB= BUA

AB= BA

• Associativity (A U B) U C = A U (B U C)
(A  B)  C = A  (B  C)

• Distributivity A U (B  C) = (A U B)  (A U C)
A  (B U C) = (A  B) U (A  C)

24
 Identities

• DeMorgan’s Law I (A U B)c = Ac  Bc

• DeMorgan’s Law II (A  B)c = Ac U Bc

Hand waving is
good for intuition,
p q but we aim for a
more formal proof.

25
 Proving identities

• Show that A  B and that A  B. New & important

• Use a membership table. Like truth tables

• Use previously proven identities. Like 

• Use logical equivalences to prove equivalent set

definitions.
Not hard, a little tedious

26
 Proving identities (1)

Prove that (A U B)c = Ac  Bc

1. () (x  (A U B)c)  (x  A U B)  (x  A and x  B)  (x  Ac  Bc)

2. () (x  Ac  Bc)  (x  A and x  B)  (x  A U B)  (x  (A U B)c)

27
 Proving identities (2)

Prove that (A U B)c = Ac  Bc using a membership table.

0 : x is not in the specified set

1 : otherwise

A B A B AB AUB AUB

1 1 0 0 0 1 0
1 0 0 1 0 1 0
0 1 1 0 0 1 0
0 0 1 1 1 0 1

Have we not seen

this before? 28
 Proving identities (3)

Prove that (A U B)c = Ac  Bc using identities.

(A U B) = A U B

=AB

=AB

29
 Proving identities (4)

Prove that (A U B)c = Ac  Bc using logically equivalent

set definitions.

(A U B)c = {x : (x  A v x  B)}

= {x : (x  A)  (x  B)}

= {x : (x  Ac)  (x  Bc)}

= Ac  Bc

30
 A proof to do

X  (Y - Z) = (X  Y) - (X  Z). True or False?

Prove your response.

(X  Y) - (X  Z) = (X  Y)  (X  Z)’

= (X  Y)  (X’ U Z’)

= (X  Y  X’) U (X  Y  Z’)

= U (X  Y  Z’)
= (X  Y  Z’)

31
 Another proof to do

AB=
Prove that if (A - B) U (B - A) = (A U B) then ______
a) AUB=
b) A=B
Suppose to the contrary, that A  B  , and that x  A  B. c) AB=
d) A-B = B-A = 

Then x cannot be in A-B and x cannot be in B-A.

DeMorgan’s Law
Then x is not in (A - B) U (B - A).

Do you see the contradiction yet?

But x is in A U B since (A  B)  (A U B). Trying to prove p  q

Assume p and not q, and find a
Thus, A  B = . contradiction.
Our contradiction was that sets were
not equal.
32
 Wrap up

• Sets are an essential structure for

all mathematics.
• We covered the basic definitions
and identities that will allow us
to reason about sets.

33