MATH2222 Homework 4

Isaac Leong
March 2023

Collaboration Statement
For this Assignment, because I started very early, I unfortunately did not have a chance to
discuss and collaborate with others in the class. I finished this assignment all the questions
on my own. However, for question 3, I have discussed with Thomas Bridgart about another
possible solution. Also, I discussed some other questions for pleasure with him during the
tutorial.

1
Question 1
(a)
Prove: The composition of injective functions is injective.
Suppose function f : X −→ Y and g : Y −→ V are injective. We will prove that the
composition g ◦ f : X −→ V is injective.
First, by definition of injection, both functions f and g satisfy the following conditions.

If f (x1 ) = f (x2 ) ,then x1 = x2

If g(y1 ) = g(y2 ) ,then y1 = y2

Then, g ◦ f (x1 ) = g ◦ f (x2 ) =⇒ g(f (x1 )) = g(f (x2 ))

=⇒ g(x1 ) = g(x2 ) (f is an injection)
=⇒ x1 = x2 (g is an injection)

Therefore, we have proven that if g ◦ f (x1 ) = g ◦ f (x2 ), then x1 = x2 .

This implies that g ◦ f is injective. ■

(b)
Prove: The composition of surjective functions is surjective.
Suppose function f : X −→ Y and g : Y −→ V are surjective. We will prove that the
composition g ◦ f : X −→ V is also surjective.
Then, by definition of surjection,

∀v ∈ V , ∃y ∈ Y such that g(y) = v (1)

∀y ∈ Y , ∃x ∈ X such that f (x) = y (2 )

Then by combining (1) and (2),

∀v ∈ V , ∃x ∈ X such that g(y) = g(f (x)) = g ◦ f (x) = v

Therefore, by definition of surjection, g ◦ f is a surjection. ■

2
Question 2
Let f : A −→ B and g : B −→ C, and let h = g ◦ f .

(a)
Prove or disprove: If h is injective, then f is injective.
Direct prove for case 1 but prove by contradiction for case 2.

Case 1:
If there are only one or no element in A, then f has to be an injection because for all elements
in B, there is only at most one element in A such that f (a) = b.

Case 2:
If there are at least two elements in set A, then we can make the following assumption.

∃a1 , a2 ∈ A such that f (a1 ) = f (a2 ) but a1 ̸= a2

Since f (a1 ) = f (a2 ), then g(f (a1 )) = g(f (a2 )), where a1 ̸= a2 . This is equivalently h(a1 ) =
h(a2 ) (a1 ̸= a2 ).
However, this contradicts to h being injective. Therefore, the assumption is wrong and f is
injective. ■

(b)
Prove or disprove: If h is injective, then g is injective.
This statement is false.
Let, A = {1}, B = R and C = R.
Let f be a constant function, ie. f (a) = 1. Then, let g(b) = b2 .
Then, since for all c ∈ R, there is only at most one value a ∈ A, such that g ◦ f (a) = c. This
implies that g ◦ f is injective. Therefore, even though g ◦ f is injective, g is not injective. ■

(c)
Prove or disprove: If h is surjective, then f is surjective.
This statement is false. Construct the following functions:

f : [−1, 1] −→ [−1, 1] and g : [−1, 1] −→ [0, 1]

f (x) = x2 and g(x) = x2

Then, g ◦ f (x) = x4
1
This function is clearly surjective. Since for all y ∈ [0, 1], we can just map it to x = y 4 ,
x ∈ [0, 1]. This shows that for all elements in the codomain, [0,1], there is at least one
element in the domain, [−1, 1], that corresponds. ■

3
(d)
Prove or disprove: If h is surjective, then g is surjective.
Proof: Since h is surjective, then for all c ∈ C, there exists a ∈ A such that h(a) = c.
Also note that h(a) = g(f (a)). Since, f (A) ⊂ B, therefore f (a) ∈ B. Therefore, we can just
choose b = f (a) ∈ B such that g(b) = c. This implies that for all c ∈ C, there exists b ∈ B
such that g(b) = c. Therefore, g is surjective. ■

4
Question 3
Let n and k be positive integers. Prove that if there is a bijection from [n] to [k] then n = k.
(Denote {1, ....., n}, by [n])

Prove by contrapositive and contradiction.

First, the contrapositive of the statement is: If n ̸= k, then there is no bijection between
[n] and [k].
Then, prove this statement by contradiction. Assume that there is still a bijection be-
tween [n] and [k] even if n ̸= k.
Case 1: k > n. Suppose the bijection is f . Then, since f is surjective, therefore for all
b ∈ [k], there is a ∈ [n] such that f (a) = b. Since each element in the domain, [n], can only
map to one single value in the codomain, [k], thee must be at least k elements in [n]. Since
by definition, [n] = {1, ...., n}, therefore n ≥ k. This leads to contradiction.
Case 2: k < n. Again suppose the bijection is f . Then, since f is injective, all elements
in the the domain, [n], map to different value in the codomain, [k]. Therefore, there should
also be n elements in f ([n]). Also, by definition of codomain, f ([n]) ⊂ [k]. Therefore, there
should be at least n elements in the set [k]. Since [k] includes all natural numbers from 1 to
k, then in order to have at least n elements, k ≥ n. This leads to contradiction as well.
Therefore, the original hypothesis is wrong. Therefore, If n ̸= k, then there is no bijection
between [n] and [k]. Therefore, it is also true that if there is a bijection from [n] to [k], then
n = k. ■

5
Question 4
Show that odd integers are a countable set.
First, to show that it is a countable set, we have to show that there is a bijection between
it and the natural number set by definition of a countable set.
Since, there is a bijection from the odd integer sets to the natural number set, if and only
if there exists a bijection from the natural number set to the odd integers set.
Hence, we construct the following bijection going from natural number to the odd integer
set.
f : N −→ {2k + 1|k ∈ Z} (N includes zero)
(
x if x is odd
f (x) =
−x − 1 if x is even
In order to show the construction is indeed a bijection, it is required to show that it is
both injective and surjective.

Injective
If f (x) > 0, then from the definition, x is odd. Then, f is f (x) = x. Therefore, f (x1 ) =
f (x2 ) =⇒ x1 = x2 .
If f (x) < 0, then from the definition, x is even. Then the function f (x) = −x − 1.
Therefore,

f (x1 ) = f (x2 ) =⇒ −x1 − 1 = −x2 − 1

=⇒ −x1 = −x2
=⇒ x1 = x2

This shows that f is also injective for all f (x) < 0. Therefore, f is injective for the domain,
N, and the codomain, {2k + 1|k ∈ Z}, provided.

Surjective
Prove by contradiction. Suppose there is an odd integer, y0 , that is not hit by any elements
in the natural number according to the mapping of f .
If y0 > 0, then since there is always x = y0 such that f (x0 ) = y by definition of the
function. Therefore, y0 is not positive.
Also, if y0 < 0, then since there is always x = −y0 − 1 such that f (x0 ) = y by definition
of the function because, if y0 is a negative odd integer, then x = −y0 − 1 is a positive even
integer (including 0). Therefore, y0 is not negative as well.
Since, y0 is an odd number, therefore, it is not zero as well. This leads to contradiction
of the law of trichotomy. Therefore, the original assumption is wrong and f should indeed
be surjective.

Therefore, in conclusion, f is a bijective mapping from N to all odd integers. Therefore,

all odd integer is a countable set. ■

6
Question 5
Let A1 , A2 , A3 , ... be a sequence of disjoint sets, each of which is countable. Prove that their
union is countable.
First, from the question, there are infinitely many sets Ai indexed by the natural number.
Then, if we denote the jth element in a set Ai as aij , then all elements in all the sets can
be indexed by (i, j), where i ∈ N and j ∈ N, where i means the element is in the i-th set,
Ai , and it is the j-th element in the set. This is equivalently indexing all elements in all the
set by an element of N × N.
A1 = {a11 , a12 , a13 , a14 , ...}
A2 = {a21 , a22 , a23 , a24 , ...}
A3 = {a31 , a32 , a33 , a34 , ...}
:
:
Then, through the bijection introduced in class,

f : N × N −→ N

f (i, j) = 2i−1 (2j − 1)

This shows that the set N × N is countable because it is a bijection between N × N and N.
Then from the discussion above, the thing that we have actually done is creating a bijection,
g, from N × N to all elements of all the sets, which can be proven in the following.

Injective
Since, all the Ai are disjoint, this implies that they do not have common element, ie. Ai ∩Aj =
∅. Therefore, if g(i, j) = aij = axy = g(x, y), then (i, j) = (x, y) because all aij are unique.
Hence, g is injective.

Surjective
Since, by definition for the mapping it is constructed above, any element in any set are
assigned a pair of integers based on which set Ai it belongs to and the place it appears in
Ai , therefore, for all aij ∈ A1 ∪ A2 ∪ A3 ..., there exists an ordered pair of (i, j) such that
g(i, j) = aij . Hence, by definition, g is surjective.

Therefore, the mapping,g, from A1 ∪ A2 ∪ A3 ... to N × N is bijective. Also, since the

mapping, f , from N × N is also bijective, then as proven in question 1, the mapping f ◦ g is
also bijective. Therefore, there is a bijective mapping from A1 ∪ A2 ∪ A3 ∪ A4 ... to N. Hence,
by definition of countability, A1 ∪ A2 ∪ A3 ∪ A4 ... is countable. ■