以我相对平庸的智商来说,Analysis I 的第8章是整本书中相对比较难的章节之一,由于(以个人的理解)这章实际是将集合论和实分析的起步阶段内容都介绍了,但是篇幅又不会展开到实分析教科书的程度,对于第一次接触的人毫无疑问是比较棘手的,我虽然之前接触过实分析,大致能理解countable和uncountable的区别,但是从没有非常仔细的思考过著名的选择公理Axiom of Choice,以及其各种等价结论,还有其应用。这一章提供了一个机会。
本章的习题也是 Analysis I 中数量比较大,难度也挺高的一部分,基本上如果没有hint,都很难做出来。这种熬脑的感觉在第1节还好,在第5节达到了顶峰。
Exercise 8.1.1
First, if there exists a proper subset
Y
⊊
X
Y⊊X
Y⊊X which has the same cardinality as
X
X
X, then assume
X
X
X is finite, then
Y
Y
Y must be finite. Since
X
=
Y
∩
(
X
\
Y
)
X=Y∩(X\backslash Y)
X=Y∩(X\Y), we know that
Y
Y
Y and
X
\
Y
X\backslash Y
X\Y are disjoint, further
X
\
Y
X\backslash Y
X\Y is not empty, so
#
(
X
\
Y
)
≥
1
\#(X\backslash Y)≥1
#(X\Y)≥1, and
#
(
X
)
=
#
(
Y
)
+
#
(
X
\
Y
)
=
#
(
X
)
+
#
(
X
\
Y
)
≥
#
(
X
)
+
1
>
#
(
X
)
\#(X)=\#(Y)+\#(X\backslash Y)=\#(X)+\#(X\backslash Y)≥\#(X)+1>\#(X)
#(X)=#(Y)+#(X\Y)=#(X)+#(X\Y)≥#(X)+1>#(X)
which is a contradiction.
On the contrary, if
X
X
X is infinite, then we can form a set
S
S
S like this:
Choose
∀
x
∈
X
∀x∈X
∀x∈X, rename it
x
0
x_0
x0 and let
S
0
=
{
x
0
}
S_0=\{x_0 \}
S0={x0}, then choose any element belongs to
X
\
S
0
X\backslash S_0
X\S0, and rename it
x
1
x_1
x1, then let
S
1
=
S
0
∪
{
x
1
}
S_1=S_0∪\{x_1 \}
S1=S0∪{x1}. This process shall never stop, since
X
X
X is infinite and we have the Axiom of Choice. Doing this recursively, we can eventually get a set
S
S
S (by letting
n
→
∞
n→∞
n→∞), obviously
S
⊆
X
S⊆X
S⊆X.
We let
Y
=
X
\
{
x
0
}
Y=X\backslash \{x_0\}
Y=X\{x0}, then
Y
Y
Y is a proper subset of
X
X
X. For any
x
∈
X
x∈X
x∈X, either
x
∈
S
x∈S
x∈S or
x
∈
X
\
S
x∈X\backslash S
x∈X\S, thus we define a function
f
f
f from
X
X
X to
Y
Y
Y as follows:
f
(
x
)
=
{
x
n
+
1
,
x
=
x
n
∈
S
x
,
x
∈
X
\
S
f(x)=\begin{cases}x_{n+1},&x=x_n∈S\\x,&x∈X\backslash S \end{cases}
f(x)={xn+1,x,x=xn∈Sx∈X\S
It’s easy to show that
f
f
f is a bijection.
Exercise 8.1.2
For the sake of contradiction we assume Proposition 8.1.4 is false, then there’s a nonempty set
X
⊆
N
X⊆\mathbf N
X⊆N, such that for any
n
∈
X
n∈X
n∈X, there exists one
m
∈
X
m∈X
m∈X s.t.
m
<
n
m<n
m<n.
Since
X
≠
∅
X≠∅
X=∅, we choose
x
∈
X
x∈X
x∈X, and let
x
=
a
0
x=a_0
x=a0, by our assumption, we can find an element
a
1
∈
X
,
a
1
<
x
=
a
0
a_1∈X,a_1<x=a_0
a1∈X,a1<x=a0, once we find
a
n
a_n
an, we can by assumption find an element
a
n
+
1
∈
X
a_{n+1}∈X
an+1∈X such that
a
n
+
1
<
a
n
a_{n+1}<a_n
an+1<an, thus we build a infinite descent sequence
(
a
n
)
n
=
0
∞
(a_n )_{n=0}^∞
(an)n=0∞ of natural numbers, this contradicts the principle of infinite descent.
If we replace the natural numbers by the integers, then the well-ordering principle doesn’t work. For example the set
X
=
{
−
1
,
−
2
,
−
3
,
…
}
X=\{-1,-2,-3,…\}
X={−1,−2,−3,…} doesn’t have a minimum element.
If we replace the natural numbers by the positive rationals, then the well-ordering principle doesn’t work. For example the set
X
=
{
q
∈
Q
:
q
>
0
}
X=\{q∈\mathbf Q:q>0\}
X={q∈Q:q>0} doesn’t have a minimum element.
Exercise 8.1.3
I will prove this Proposition in a complete manner.
Recursively define a sequence of natural numbers
a
0
,
a
1
,
a
2
,
…
a_0,a_1,a_2,…
a0,a1,a2,… using the formula below:
a
n
≔
min
{
x
∈
X
:
∀
m
<
n
,
x
≠
a
m
}
a_n≔\min \{x∈X:∀m<n,x≠a_m \}
an:=min{x∈X:∀m<n,x=am}
Since
X
X
X is infinite, the set
{
x
∈
X
:
∀
m
<
n
,
x
≠
a
m
}
\{x∈X:∀m<n,x≠a_m \}
{x∈X:∀m<n,x=am} should also be infinite, assume not, then for some n this set is finite, and
{
a
0
,
a
1
,
…
,
a
n
−
1
}
\{a_0,a_1,…,a_{n-1} \}
{a0,a1,…,an−1} is finite, by
X
=
{
x
∈
X
:
∀
m
<
n
,
x
≠
a
m
}
∪
{
a
0
,
a
1
,
…
,
a
n
−
1
}
X=\{x∈X:∀m<n,x≠a_m \}∪\{a_0,a_1,…,a_{n-1} \}
X={x∈X:∀m<n,x=am}∪{a0,a1,…,an−1}
We deduce
X
X
X is finite, which is a contradiction.
Since
{
x
∈
X
:
∀
m
<
n
,
x
≠
a
m
}
\{x∈X:∀m<n,x≠a_m \}
{x∈X:∀m<n,x=am} is infinite, it’s not empty for every
n
n
n, so we can successfully define
min
{
x
∈
X
:
∀
m
<
n
,
x
≠
a
m
}
\min\{x∈X:∀m<n,x≠a_m \}
min{x∈X:∀m<n,x=am} by the well-ordering principle.
For
∀
n
∀n
∀n, we know that
a
n
+
1
≔
min
{
x
∈
X
:
∀
m
<
n
+
1
,
x
≠
a
m
}
a_{n+1}≔\min\{x∈X:∀m<n+1,x≠a_m \}
an+1:=min{x∈X:∀m<n+1,x=am}
By the definition of
a
n
+
1
a_{n+1}
an+1 we can see
a
n
+
1
≠
a
n
a_{n+1}≠a_n
an+1=an and
a
n
+
1
∈
{
x
∈
X
:
∀
m
<
n
,
x
≠
a
m
}
a_{n+1}∈\{x∈X:∀m<n,x≠a_m \}
an+1∈{x∈X:∀m<n,x=am}, since
a
n
≔
min
{
x
∈
X
:
∀
m
<
n
,
x
≠
a
m
}
a_n≔\min\{x∈X:∀m<n,x≠a_m \}
an:=min{x∈X:∀m<n,x=am}, we can deduce
a
n
<
a
n
+
1
,
∀
n
a_n<a_{n+1},∀n
an<an+1,∀n. Thus
a
0
<
a
1
<
a
2
<
⋯
a_0<a_1<a_2<⋯
a0<a1<a2<⋯
Which means for all
n
≠
m
,
a
n
≠
a
m
n≠m,a_n≠a_m
n=m,an=am, otherwise we would get
a
n
<
a
n
a_n<a_n
an<an or
a
m
<
a
m
a_m<a_m
am<am. Also we know by the definition of
min
(
X
)
,
a
n
∈
X
,
∀
n
\min(X), a_n∈X,∀n
min(X),an∈X,∀n.
Now we define
f
:
N
→
X
f:\mathbf N→X
f:N→X by
f
(
n
)
≔
a
n
f(n)≔a_n
f(n):=an, we already show
f
f
f is 1-to-1. To prove
f
f
f is surjective, choose
x
∈
X
x∈X
x∈X, assume no
n
n
n can let
f
(
n
)
=
x
f(n)=x
f(n)=x, then this means
a
n
≠
x
,
∀
n
a_n≠x,∀n
an=x,∀n, so for every
n
,
x
∈
{
x
∈
X
:
∀
m
<
n
,
x
≠
a
m
}
n, x∈\{x∈X:∀m<n,x≠a_m \}
n,x∈{x∈X:∀m<n,x=am}, thus
x
≥
a
n
,
∀
n
x≥a_n,∀n
x≥an,∀n. As
(
a
n
)
n
=
0
∞
(a_n )_{n=0}^∞
(an)n=0∞ is an increasing sequence in natural numbers, we can prove by induction that
a
n
≥
n
a_n≥n
an≥n, thus
x
≥
n
x≥n
x≥n for every
n
n
n, we can then deduce
x
≥
x
+
1
x≥x+1
x≥x+1 and get a contradiction. So
f
f
f is surjective, thus bijective.
Last, we assume there’s another increasing bijective
g
:
N
→
X
,
g
≠
f
g:\mathbf N→X,g≠f
g:N→X,g=f, then
{
n
∈
N
:
g
(
n
)
≠
f
(
n
)
}
\{n∈\mathbf N:g(n)≠f(n)\}
{n∈N:g(n)=f(n)} is not empty, which means we can define
m
≔
min
{
n
∈
N
:
g
(
n
)
≠
f
(
n
)
}
m≔\min\{n∈\mathbf N:g(n)≠f(n)\}
m:=min{n∈N:g(n)=f(n)}
So
g
(
m
)
≠
f
(
m
)
=
a
m
g(m)≠f(m)=a_m
g(m)=f(m)=am, and for all
n
<
m
,
g
(
n
)
=
f
(
n
)
=
a
n
n<m,g(n)=f(n)=a_n
n<m,g(n)=f(n)=an, in that case,
g
(
m
)
g(m)
g(m) must equal some
x
∈
X
x∈X
x∈X, since
g
g
g is increasing, we have
x
>
g
(
n
)
,
∀
n
<
m
x>g(n),∀n<m
x>g(n),∀n<m, in particular:
x
∈
{
x
∈
X
:
∀
n
<
m
,
x
≠
a
n
}
⇒
x
≥
min
{
x
∈
X
:
∀
n
<
m
,
x
≠
a
n
}
=
a
m
x∈\{x∈X:∀n<m,x≠a_n \}⇒x≥\min\{x∈X:∀n<m,x≠a_n \}=a_m
x∈{x∈X:∀n<m,x=an}⇒x≥min{x∈X:∀n<m,x=an}=am
Together with
g
(
m
)
≠
a
m
g(m)≠a_m
g(m)=am we know that
g
(
m
)
>
a
m
∈
X
g(m)>a_m∈X
g(m)>am∈X, since
g
g
g is strictly increasing, this means
a
m
∉
g
(
N
)
a_m∉g(\mathbf N)
am∈/g(N), or
g
g
g is not surjective, a contradiction.
Exercise 8.1.4
We define
A
A
A as what Hint suggests. Consider
f
∣
A
:
A
→
f
(
N
)
f|_A:A→f(\mathbf N)
f∣A:A→f(N), we prove
f
∣
A
f|_A
f∣A is a bijection.
Let
m
,
n
∈
A
,
m
≠
n
m,n∈A,m≠n
m,n∈A,m=n, then we shall have
m
<
n
m<n
m<n or
n
<
m
n<m
n<m, in either case, we can deduce from the definition of
A
A
A that
f
∣
A
(
m
)
≠
f
∣
A
(
n
)
f|_A (m)≠f|_A (n)
f∣A(m)=f∣A(n), so
f
∣
A
f|_A
f∣A is injective.
For
∀
y
∈
f
(
N
)
∀y∈f(\mathbf N)
∀y∈f(N), we can have
k
∈
N
k∈\mathbf N
k∈N, s.t.
y
=
f
(
k
)
y=f(k)
y=f(k). Now for all
0
≤
n
≤
k
0≤n≤k
0≤n≤k, we can examine
f
(
n
)
f(n)
f(n) recursively starting from
0
0
0 and choose the first
m
m
m such that
f
(
m
)
=
y
f(m)=y
f(m)=y. If
m
=
k
m=k
m=k, this means
k
∈
A
k∈A
k∈A. If
m
<
k
m<k
m<k, then
m
∈
A
m∈A
m∈A and
f
(
m
)
=
y
f(m)=y
f(m)=y. In either case we have
y
∈
f
∣
A
(
A
)
y∈f|_A (A)
y∈f∣A(A), so
f
∣
A
f|_A
f∣A is surjective.
Since
A
A
A is a subset of
N
\mathbf N
N and is at most countable,
f
(
N
)
f(\mathbf N)
f(N) is at most countable.
Exercise 8.1.5
X
X
X is countable, thus we can find a bijection
g
:
N
→
X
g:\mathbf N→X
g:N→X, then
h
=
f
∘
g
h=f∘g
h=f∘g is a function from
N
\mathbf N
N to
Y
Y
Y, thus use Proposition 8.1.8,
f
(
X
)
=
f
(
g
(
N
)
)
=
h
(
N
)
f(X)=f(g(\mathbf N))=h(\mathbf N)
f(X)=f(g(N))=h(N) is at most countable.
Exercise 8.1.6
If there exists an injective map
f
:
A
→
N
f:A→\mathbf N
f:A→N, then
f
f
f is a bijection from
A
A
A to
f
(
A
)
f(A)
f(A), since
f
(
A
)
f(A)
f(A) is a subset of
N
\mathbf N
N and is at most countable,
A
A
A is at most countable as
f
−
1
f^{-1}
f−1 is a function from
f
(
A
)
f(A)
f(A) to
A
A
A and
A
=
f
−
1
(
f
(
A
)
)
A=f^{-1} (f(A))
A=f−1(f(A)) and use Corollary 8.1.9.
If
A
A
A is at most countable, then
A
A
A is finite or
A
A
A is countable. Then
For the case
A
A
A is finite, we can see
A
A
A has cardinality
n
n
n for some natural number
n
n
n. We select a bijection
g
g
g from
A
A
A to
{
1
,
2
,
…
,
n
}
\{1,2,…,n\}
{1,2,…,n}, and define
f
:
A
→
N
f:A→\mathbf N
f:A→N by
f
(
a
)
=
g
(
a
)
,
∀
a
∈
A
f(a)=g(a),∀a∈A
f(a)=g(a),∀a∈A, then
f
f
f is injective since
g
g
g is bijective.
For the case
A
A
A is countable, we can find a bijection
f
:
A
→
N
f: A→\mathbf N
f:A→N, this function is injective.
Exercise 8.1.7
We use all the notations from Hint, then
h
:
N
→
X
∪
Y
h:\mathbf N→X∪Y
h:N→X∪Y is defined.
We need to show
h
(
N
)
=
X
∪
Y
h(\mathbf N)=X∪Y
h(N)=X∪Y, notice first that
∀
n
∈
N
,
h
(
n
)
∈
X
∀n∈\mathbf N,h(n)∈X
∀n∈N,h(n)∈X if
n
n
n is even, and
h
(
n
)
∈
Y
h(n)∈Y
h(n)∈Y if
n
n
n is odd, thus
h
(
n
)
∈
X
∪
Y
,
∀
n
∈
N
h(n)∈X∪Y,∀n∈\mathbf N
h(n)∈X∪Y,∀n∈N, or
h
(
N
)
⊆
X
∪
Y
h(\mathbf N)⊆X∪Y
h(N)⊆X∪Y.
Next, we have
(
x
∈
X
∪
Y
)
⇒
(
x
∈
X
)
∨
(
x
∈
Y
)
⇒
(
x
=
f
(
m
)
)
∨
(
x
=
g
(
n
)
)
⇒
(
x
=
h
(
2
m
)
)
∨
(
x
=
h
(
2
n
+
1
)
)
⇒
(
x
∈
h
(
N
)
)
\begin{aligned}(x∈X∪Y)&⇒(x∈X)∨(x∈Y)⇒(x=f(m))∨(x=g(n))\\&⇒(x=h(2m))∨(x=h(2n+1))⇒(x∈h(\mathbf N))\end{aligned}
(x∈X∪Y)⇒(x∈X)∨(x∈Y)⇒(x=f(m))∨(x=g(n))⇒(x=h(2m))∨(x=h(2n+1))⇒(x∈h(N))
Thus
X
∪
Y
⊆
h
(
N
)
X∪Y⊆h(\mathbf N)
X∪Y⊆h(N). Then we can have
X
∪
Y
=
h
(
N
)
X∪Y=h(\mathbf N)
X∪Y=h(N)
By Corollary 8.1.9 we know that
X
∪
Y
X∪Y
X∪Y is at most countable, we remains to show
X
∪
Y
X∪Y
X∪Y is not finite. Assume so, then
X
∪
Y
X∪Y
X∪Y has cardinality
n
n
n for some natural number
n
n
n. So
X
⊆
X
∪
Y
⇒
#
(
X
)
≤
n
X⊆X∪Y ⇒ \#(X)≤n
X⊆X∪Y⇒#(X)≤n
But if we let
S
=
{
1
,
2
,
…
,
n
+
1
}
S=\{1,2,…,n+1\}
S={1,2,…,n+1}, then
f
:
S
→
f
(
S
)
f:S→f(S)
f:S→f(S) is a bijection, thus
f
(
S
)
f(S)
f(S) has cardinality
n
+
1
n+1
n+1, however,
f
(
S
)
⊆
X
f(S)⊆X
f(S)⊆X, so
#
(
X
)
≥
n
+
1
\#(X)≥n+1
#(X)≥n+1, this is a contradiction.
Exercise 8.1.8
By hypothesis, we have a bijection
f
:
N
→
X
f:\mathbf N→X
f:N→X and a bijection
g
:
N
→
Y
g:\mathbf N→Y
g:N→Y. We define a function
h
:
N
×
N
→
X
×
Y
h:\mathbf N×\mathbf N→X×Y
h:N×N→X×Y by
h
(
m
,
n
)
=
(
f
(
m
)
,
g
(
n
)
)
h(m,n)=(f(m),g(n))
h(m,n)=(f(m),g(n)), then we have
(
h
(
m
,
n
)
=
h
(
m
′
,
n
′
)
)
⇒
(
(
f
(
m
)
=
f
(
m
′
)
)
∧
(
g
(
n
)
=
g
(
n
′
)
)
)
⇒
(
m
=
m
′
)
∧
(
n
=
n
′
)
⇒
(
m
,
n
)
=
(
m
′
,
n
′
)
\begin{aligned}(h(m,n)=h(m’,n’))&⇒((f(m)=f(m’ ))∧(g(n)=g(n’ )))\\&⇒(m=m’ )∧(n=n’ )⇒(m,n)=(m’,n’ )\end{aligned}
(h(m,n)=h(m′,n′))⇒((f(m)=f(m′))∧(g(n)=g(n′)))⇒(m=m′)∧(n=n′)⇒(m,n)=(m′,n′)
Thus
h
h
h is injective. And for
∀
(
x
,
y
)
∈
X
×
Y
∀(x,y)∈X×Y
∀(x,y)∈X×Y, we have
x
∈
X
x∈X
x∈X and
y
∈
Y
y∈Y
y∈Y, so
∃
m
,
n
∈
N
∃m,n∈N
∃m,n∈N, s.t.
f
(
m
)
=
x
,
g
(
n
)
=
y
f(m)=x,g(n)=y
f(m)=x,g(n)=y, thus
h
(
m
,
n
)
=
(
x
,
y
)
h(m,n)=(x,y)
h(m,n)=(x,y), which means
h
h
h is surjective.
Now that
h
h
h is a bijection, from Corollary 8.1.13 we know
X
×
Y
X×Y
X×Y is countable.
Exercise 8.1.9
If
I
I
I is finite and
A
α
A_α
Aα is finite for every
α
α
α, then the statement is obvious.
Now suppose
I
I
I is countable, then we have a bijection
g
:
N
→
I
g:\mathbf N→I
g:N→I, for each
α
∈
I
α∈I
α∈I such that
A
α
≠
∅
A_α≠∅
Aα=∅, we assign a function
f
α
:
N
→
A
α
f_α:\mathbf N→A_α
fα:N→Aα as follows:
If
A
α
A_α
Aα is finite, then it’s possible to write
A
α
=
{
x
1
,
…
,
x
n
}
,
n
∈
N
A_α=\{x_1,…,x_n \},n∈\mathbf N
Aα={x1,…,xn},n∈N, then let
f
α
(
i
)
=
{
x
i
,
1
≤
i
≤
n
x
1
,
i
>
n
f_α (i)=\begin{cases}x_i,&1≤i≤n\\x_1,&i>n\end{cases}
fα(i)={xi,x1,1≤i≤ni>n
If
A
α
A_α
Aα is infinite or countable, then we can find a bijection
f
′
:
N
→
A
α
f’:\mathbf N→A_α
f′:N→Aα, we let
f
α
=
f
′
f_α=f’
fα=f′.
So as long as
A
α
A_α
Aα is nonempty, we have assigned a function
f
α
:
N
→
A
α
f_α:\mathbf N→A_α
fα:N→Aα. Next, if denote
J
=
{
α
∈
I
:
A
α
≠
∅
}
J=\{α∈I:A_α≠∅\}
J={α∈I:Aα=∅} we define
h
:
N
×
N
→
⋃
α
∈
J
A
α
h: N×N→⋃_{α∈J}A_α
h:N×N→⋃α∈JAα as follows:
h
(
m
,
n
)
=
f
g
(
m
)
(
n
)
h(m,n)=f_{g(m)} (n)
h(m,n)=fg(m)(n)
Then use Corollary 8.1.13 and Corollary 8.1.9, we can see that
⋃
α
∈
J
A
α
⋃_{α∈J}A_α
⋃α∈JAα is at most countable. The final statement results from
⋃
α
∈
J
A
α
=
⋃
α
∈
I
A
α
⋃_{α∈J}A_α =⋃_{α∈I}A_α
⋃α∈JAα=⋃α∈IAα .
Exercise 8.1.10
We define
Q
n
=
{
a
/
b
:
∣
a
∣
+
b
≤
n
,
a
∈
Z
,
b
∈
N
+
}
Q_n=\{a/b:|a|+b≤n,a∈\mathbf Z,b∈\mathbf N^+\}
Qn={a/b:∣a∣+b≤n,a∈Z,b∈N+} for every
n
∈
N
n∈\mathbf N
n∈N, then
Q
0
=
∅
,
Q
1
=
{
0
}
,
Q
2
=
{
0
,
1
,
−
1
}
Q_0=∅,Q_1=\{0\}, Q_2=\{0,1,-1\}
Q0=∅,Q1={0},Q2={0,1,−1}, etc. Each
Q
n
Q_n
Qn is finite, thus have cardinality
c
n
c_n
cn, we can see
c
0
=
0
,
c
1
=
1
,
c
2
=
3
c_0=0,c_1=1,c_2=3
c0=0,c1=1,c2=3, etc. As
Q
n
⊆
Q
n
+
1
Q_n⊆Q_{n+1}
Qn⊆Qn+1, we define
S
n
=
Q
n
\
(
⋃
i
=
0
n
−
1
Q
i
)
,
n
∈
N
+
S_n=Q_n\backslash \left(⋃_{i=0}^{n-1}Q_i \right),\quad n∈\mathbf N^+
Sn=Qn\(i=0⋃n−1Qi),n∈N+
since
1
n
−
1
∈
Q
n
\frac{1}{n-1}∈Q_n
n−11∈Qn, but
1
n
−
1
∉
⋃
i
=
0
n
−
1
Q
i
\frac{1}{n-1}∉⋃_{i=0}^{n-1}Q_i
n−11∈/⋃i=0n−1Qi , each
S
n
S_n
Sn is nonempty for
n
>
1
n>1
n>1, and
S
1
=
Q
1
=
{
0
}
S_1=Q_1=\{0\}
S1=Q1={0}, so each
S
n
S_n
Sn is nonempty for
n
∈
N
+
n∈\mathbf N^+
n∈N+.
Now we define
q
n
=
c
n
−
c
n
−
1
q_n=c_n-c_{n-1}
qn=cn−cn−1, then each
S
n
S_n
Sn has cardinality
q
n
q_n
qn, so
(
q
n
)
n
=
1
∞
(q_n )_{n=1}^∞
(qn)n=1∞ is a positive increasing sequence.
We define a function
f
:
N
→
Q
f:\mathbf N→\mathbf Q
f:N→Q inductively as follows:
Start from
0
0
0, we let
f
1
(
0
)
=
0
f_1 (0)=0
f1(0)=0, which is in fact a bijection from
{
i
∈
N
:
c
0
≤
i
<
c
1
}
\{i∈\mathbf N:c_0≤i<c_1\}
{i∈N:c0≤i<c1} to
S
1
S_1
S1.
Then for any
S
n
S_n
Sn, we can find a bijection
g
n
:
{
i
∈
N
:
0
≤
i
<
q
n
}
→
S
n
g_n:\{i∈\mathbf N:0≤i<q_n \}→S_n
gn:{i∈N:0≤i<qn}→Sn, we then define
f
n
(
i
)
=
g
n
(
i
−
c
n
−
1
)
,
c
n
−
1
≤
i
<
c
n
f_n (i)=g_n (i-c_{n-1} ),\quad c_{n-1}≤i<c_n
fn(i)=gn(i−cn−1),cn−1≤i<cn
So we know
f
n
f_n
fn is a bijection from
{
i
∈
N
:
c
n
−
1
≤
i
<
c
n
}
\{i∈\mathbf N:c_{n-1}≤i<c_n\}
{i∈N:cn−1≤i<cn} to
S
n
S_n
Sn.
Finally we can define
f
(
k
)
=
f
n
(
k
)
f(k)=f_n (k)
f(k)=fn(k), in which
n
n
n is the positive natural number which satisfies
c
n
−
1
≤
k
<
c
n
c_{n-1}≤k<c_n
cn−1≤k<cn, since if
m
≠
n
m≠n
m=n, we have
{
i
∈
N
:
c
n
−
1
≤
i
<
c
n
}
∩
{
i
∈
N
:
c
m
−
1
≤
i
<
c
m
}
=
∅
\{i∈\mathbf N:c_{n-1}≤i<c_n \}∩\{i∈\mathbf N:c_{m-1}≤i<c_m \}=∅
{i∈N:cn−1≤i<cn}∩{i∈N:cm−1≤i<cm}=∅
Since any for natural number
n
n
n, we have
{
0
,
1
,
…
,
n
−
1
}
⊆
Q
n
\{0,1,…,n-1\}⊆Q_n
{0,1,…,n−1}⊆Qn, so
c
n
≥
n
c_n≥n
cn≥n, which means
⋃
n
=
1
∞
{
i
∈
N
:
c
n
−
1
≤
i
<
c
n
}
=
N
⋃_{n=1}^∞\{i∈\mathbf N:c_{n-1}≤i<c_n\}=\mathbf N
n=1⋃∞{i∈N:cn−1≤i<cn}=N
Thus
f
f
f is truly a function defined on
N
\mathbf N
N, and the injectivity of
f
f
f is clear from the induction process, as we form
f
f
f from a sequence of bijective functions on disjoint subsets of
N
\mathbf N
N. We only remains to show
f
f
f is surjective. Choose any
q
∈
Q
q∈\mathbf Q
q∈Q, then we can write
q
=
a
/
b
q=a/b
q=a/b, in which
b
b
b is a positive natural number, we can be sure that
q
∈
Q
∣
a
∣
+
b
q∈Q_{|a|+b}
q∈Q∣a∣+b, so
q
∈
S
m
,
0
<
m
≤
∣
a
∣
+
b
q∈S_m,\quad 0<m≤|a|+b
q∈Sm,0<m≤∣a∣+b
Thus there’s a number
k
∈
{
i
∈
N
:
c
m
−
1
≤
i
<
c
m
}
k∈\{i∈\mathbf N:c_{m-1}≤i<c_m\}
k∈{i∈N:cm−1≤i<cm} such that
f
m
(
k
)
=
f
(
k
)
=
q
f_m (k)=f(k)=q
fm(k)=f(k)=q. This means
f
f
f is surjective.