Sets

A set is a collection of objects. These objects can be anything, even sets themselves. An
object x that is in a set S is called an element of that set. We write x ∈ S.

Some special sets:

∅: The empty set; the set containing no objects at all

N: The set of all natural numbers 1, 2, 3, … (sometimes, we include 0 as a natural

number)

Z: the set of all integers …, -3, -2, -1, 0, 1, 2, 3, …

Q: the set of all rational numbers (numbers that can be written as a ratio p/q with
p and q integers)

R: the set of real numbers (points on the continuous number line; comprises the
rational as well as irrational numbers)

When all elements of a set A are elements of set B as well, we say that A is a subset of B.
We write this as A ⊆ B.

When A is a subset of B, and there are elements in B that are not in A, we say that A is a
strict subset of B. We write this as A ⊂ B.

When two sets A and B have exactly the same elements, then the two sets are the same.
We write A = B.

Some Theorems:

For any sets A, B, and C:

A⊆A

A = B if and only if A ⊆ B and B ⊆ A

If A ⊆ B and B ⊆ C then A ⊆ C

Formalizations in first-order logic:

∀x ∀y (x = y ↔ ∀z (z ∈ x ↔ z ∈ y)) Axiom of Extensionality

∀x ∀y (x ⊆ y ↔ ∀z (z ∈ x → z ∈ y)) Definition subset

Operations on Sets

With A and B sets, the following are sets as well:

The union A ∪ B, which is the set of all objects that are in A or in B (or both)

The intersection A ∩ B, which is the set of all objects that are in A as well as B

The difference A \ B, which is the set of all objects that are in A, but not in B

The powerset P(A) which is the set of all subsets of A.

The Cartesian product A × B, which is the set of all ordered pairs (2-tuples) <a,
b> where a ∈ A and b ∈ B.
This generalizes to any number of sets, i.e. A1 × A2 × … × An is the set of all n-
tuples <a1, a2, …, an> where a1 ∈ A1, a2 ∈ A2, … , and an ∈ An.

Some theorems:

P(∅) = {∅}

For any sets A, B, and C:

A∪B=B∪A Commutation ∪

A∩B=B∩A Commutation ∩

A ∪ (B ∪ C) = A ∪ (B ∪ C) Association ∪

A ∩ (B ∩ C) = A ∩ (B ∩ C) Association ∩

A ∪ (B ∩ C) = (A ∪ B) ∩ (A ∪ C) Distribution ∪ over ∩

A ∩ (B ∪ C) = (A ∩ B) ∪ (A ∩ C) Distribution ∩ over ∪

Formalization in FOL:

∀x ∀y (x ∈ pow(y) ↔ x ⊆ y) Definition powerset

∀x ¬x ∈ e Definition empty set

∀x ∀y ∀z (z ∈ union(x,y) ↔ (z ∈ x ∨ z ∈ y)) Definition union

∀x ∀y ∀z (z ∈ int(x,y) ↔ (z ∈ x ∧ z ∈ y)) Definition intersection

Exercise: formalize the difference operation (use dif(x,y))

Relations

A relation over sets A1, A2, … , and An is a subset of A1 × A2 × … × An

A relation over two sets, A and B, is called a binary relation. For a binary relation R we
often write aRb instead of <a,b> ∈ R. A binary relation R over A and B is:

Left-Total iff for all a ∈ A there exists at least one b ∈ B such that aRb

Right-Total iff for all b ∈ B there exists at least one a ∈ A such that aRb

Right-Unique iff for all a ∈ A there exists at most one b ∈ B such that aRb

Left-Unique iff for all b ∈ B there exists at most one a ∈ A such that aRb

A binary relation that is both left-unique and right-unique is a one-to-one relation

A binary relation that is both left-total and right-total is a correspondence

R is an (endo-)relation on A if R is a binary relation over A and A. A relation R on A is:

Reflexive iff aRa for all a ∈ A

Non-Reflexive iff not reflexive

Irreflexive iff not aRa for all a ∈ A

Symmetric iff for all a ∈ A and b ∈ A: if aRb then bRa

Non-Symmetric iff not symmetric

Asymmetric iff for all a ∈ A and b ∈ A: if aRb then not bRa

Anti-symmetric iff for all a ∈ A and b ∈ A where a ≠ b: if aRb then not bRa
(or, what is the same thing: if aRb and bRa then a = b

Transitive iff for all a ∈ A, b ∈ A, and c ∈ A: if aRb and bRc then aRc

Non-Transitive iff not transitive

Antitransitive iff for all a ∈ A, b ∈ A, and c ∈ A: if aRb and bRc then not aRc

Theorem: A relation R is asymmetric if and only if R is anti-symmetric and irreflexive

Exercise: Prove this (using non-formal, but still proper mathematical proof)
Formalization in FOL:

∀x ∃y R(x,y) R is Left-total

∀y ∃x R(x,y) R is Right-total

∀x ∀y ∀z ((R(x,y) ∧ R(x,z)) → y = z) R is Right-unique

∀x ∀y ∀z ((R(y,x) ∧ R(z,x)) → y = z) R is Left-unique

∀x R(x,x) R is reflexive

∀x ¬R(x,x) R is irreflexive

∀x ∀y (R(x,y) → R(y,x)) R is symmetric

∀x ∀y (R(x,y) → ¬R(y,x)) R is asymmetric

∀x ∀y ((R(x,y) ∧ R(y,x)) → x = y) R is anti-symmetric

∀x ∀y ∀z ((R(x,y) ∧ R(y,z)) → R(x,z)) R is transitive

∀x ∀y ∀z ((R(x,y) ∧ R(y,z)) → ¬R(x,z)) R is intransitive

∀x ∀y (R(x,y) ∨ R(y,x)) R is total

∀x ∀y (¬x = y → (R(x,y) ∨ R(y,x)) R is connex

Some special kinds of relations:

A reflexive, symmetric, and transitive relation is an equivalence relation

A reflexive, anti-symmetric, and transitive relation is a partial order

A total, anti-symmetric, and transitive relation is a total (or linear) order

An irreflexive, anti-symmetric, and transitive relation is a strict partial order

A connex, irreflexive, anti-symmetric, and transitive relation is a strict total order

Examples:

⊆ on sets is a partial (but non-strict) order, ⊂ is partial and strict

≤ on numbers is total (but non-strict) order, < is a strict total order

Exercise: Show that all total orders are partial orders.

Functions

A function f : A → B is a binary relation over A and B that is right-unique.

(For this reason, a binary relation that is right-unique is often called functional).

A is called the domain, and B the co-domain, of the function f (note: many
mathematicians use the word domain to refer to the set of objects for which a function-
value is defined. We can call that the domain of definition. In the context of our course,
however, it is more useful to regard the domain as the domain (or universe) of discourse.)

If there is an object b such that <a, b>∈ f, then f(a) is defined as b, i.e. f(a) = b

If there is no object b such that <a, b>∈ f, then f(a) is undefined

Frequently, the domain A is a Cartesian product A1 × A2 × … × An. Accordingly, we will

write f : A1 × A2 × … × An → B and f(a1, a2, … , an) instead of f(<a1, a2, … , an>)

The range of a function f : A → B is the set of all b ∈ B for which there exists some a ∈
A such that f(a) = b

A function f : A → B is:

Total iff f(a) is defined for all a ∈ A

(Thus, a total function is a right-unique, left-total binary relation)

Partial iff f is not total

(Thus, a total function is a right-unique, but not left-total binary relation)

Surjective (or onto) iff for all b ∈ B, there exists some a ∈ A such that f(a) = b

(Thus, a surjective function is a right-unique, right-total binary relation. A

right-total binary relation is sometimes called a surjective relation. Also,
note that a function is surjective iff its range equals its co-domain)

Injective (or one-to-one) iff there do not exist a1 ∈ A and a2 ∈ A such that a1 ≠
a2 and f(a1) = f(a2)

(Thus, an injective function is a right-unique, left-unique binary relation.

A left-unique binary relation is sometimes called an injective relation.)

Bijective iff f is surjective and injective

(So this is a right-unique, right-total, left-unique binary relation.)

 A one–to-one correspondence between A and B iff f is total and bijective

(So this is a binary relation that has all 4 interesting properties: right-
unique, right-total, left-unique, and left-total. Conceptually, this means
that every element from A corresponds to exactly one element in B, and
vice versa. Indeed, the name makes sense: remember that a binary relation
that is left-total and right-total is a correspondence, i.e. this is where there
is at least one element from B is associated for every element from A, and
vice versa. Accordingly, a functional correspondence would be a function
(right-unique) that is total (left-total) and onto (right-total). Add one-to-
one (left-unique), and you not only have a functional one-to-one
correspondence, but also a binary relation that is right-unique, right-total,
left-unique, and left-total.

Somewhat more confusingly, a right-unique, right-total, left-unique, left-

total binary relation is sometimes called a bijective relation. This seems
confusing, because if for functions (right-unique), bijective simply means
surjective (right-total) and injective (left-unique), where does the left-total
property suddenly come from? Well, when mathematicians talk about
functions, they often suppose them to be total (left-total!), unless explicitly
specified to be partial. Indeed, when mathematicians talk about a bijective
function, they often mean a one-to-one correspondence. Likewise,
mathematicians often refer to a one-to-one correspondence as simply a
correspondence.)

For a one-to-one function function f : A → B, we can define its inverse function f-1 : B →
A as follows:

f-1(b) = a if f(a) = b
f-1(b) = undefined otherwise

Note that since f is one-to-one, there can at most be one a such that f(a) = b, so this
function is well-defined.

For any two functions f : A → B and g : B → C we can define the composite function (or
their composition) gf : A → C as follows:

gf(a) = g(f(a)) if f(a) is defined

gf(a) = undefined otherwise

Exercise: Show that if f is a one-to-one function, and f-1its inverse:

f is total if and only if f-1is onto

f is onto if and only if f-1is total