16

I would like to know if $1/4$ is in the Cantor set, I tried a lot without success, I need some hint or help how to proceed in this case. Maybe there is some tool or trick I can use.

Thanks a lot

user42912
  • 23,582
  • 5
    Have you tried writing it in base 3? What happens? – Giuseppe Negro Jan 29 '13 at 15:39
  • 5
    Hint: prove something more (which turns out to be easier to prove), namely prove $1/4$ and $3/4$ are both in the Cantor set. First prove they are both in $[0,1]$, then prove that if they are both in one stage then they are both in the next stage, where the stages are used in construction of the Cantor set. – GEdgar Jan 29 '13 at 15:40
  • http://en.wikipedia.org/wiki/Cantor_set#Composition –  Jan 29 '13 at 15:41
  • @GiuseppeNegro what do you mean by base 3? thank you for your comment – user42912 Jan 29 '13 at 15:49
  • with respect to the given answers, see, e.g., this post, which discusses the expression of numbers between 0 and 1 in their ternary (base three) expansion. – amWhy Jan 29 '13 at 15:52
  • @amWhy thank you for this post, it clarifies a lot – user42912 Jan 29 '13 at 16:46
  • @amWhy do you know where can I find a solution of the exercise of this post? – user42912 Jan 29 '13 at 16:48
  • @user42912 I've got a solution I worked out...I'll look around to see if I can find it somewhere on-line. If not, I'll try and upload my solution or find a way to make my solution accessible. – amWhy Jan 29 '13 at 16:51
  • @amWhy thank you a lot, this exercise is really helpful. – user42912 Jan 29 '13 at 16:56
  • To the closers: Is it necessary to close a 3-year-old question with some good answers? Wouldn't the proper thing be to protect it, so that new 1-point users cannot provide additional answers? – GEdgar Sep 14 '16 at 12:56

5 Answers5

32

$1/4$ is in the Cantor set. It is in the lower third. And it is in the upper third of the lower third. And in the lower third of that, and in the upper third of that, and so on.

The quickest way to see this is that it is exactly $1/4$ of the way from $1/3$ down to $0$, and then use self-similarity and symmetry.

A more plodding way to show it is to look at the series $$ \frac29 + \frac{2}{9^2}+\frac{2}{9^3}+\cdots=\frac14. $$ This shows that the base-$3$ expansion of $1/4$ is $0.02020202\ldots$. Since it has a base-$3$ expansion with only $0$s and $2$s, it is in the Cantor set.

24

Hint: write $\frac{1}{4}$ in base $3$.

To do this, just observe that

$$\frac{1}{4}=\frac{2}{9-1}=\frac{2}{9}\frac{1}{1-\frac{1}{9}}$$

and write the geometric series which has that limit.

N. S.
  • 132,525
10

We use mathematical induction to show $\frac{1}{4}$ and $\frac{3}{4}$ are in $C_n,\, \forall n\in \mathbb{N}\cup \{0\}$.

Clearly, they are in $[0,1]$.

Suppose they are in $C_N$. We observe the structure of $[0,\frac{1}{3}]$ after $n+1$ cuts is similar to $[0,1]$ after $n$ cuts. Hence if $x \in C_N$, then $\frac{x}{3} \in C_{N+1}$. Since by assumption, $\frac{3}{4} \in C_{N}$, we have $\frac{1}{4} \in C_{N+1}$. Moreover, since the cut is symmetry on $[0,1]$, we know $\frac{3}{4} \in C_{N+1}$.

Hence by induction, $\frac{1}{4}$ and $\frac{3}{4}$ are in $C_n,\, \forall n\in \mathbb{N}$, we have $\frac{1}{4}$ and $\frac{3}{4}$ are in $C$.

John
  • 13,204
7

From the construction of the Cantor set using "deleted thirds", one sees that each of the numbers $$\textstyle a_1={1\over3},\ \ a_2={1\over3}-{1\over9},\ \ a_3={1\over3}-{1\over9}+{1\over27}, \ \ \ldots$$ is an element of the Cantor set.

Since the sequence $(a_k)$ is the sequence of partial sums of the convergent Geometric series $\sum\limits_{n=1}^\infty(-1)^{n+1} ({1\over3^n}) $, it follows that
$$\lim_{k\rightarrow\infty} a_k= \sum_{n=1}^\infty(-1)^{n+1} ({\textstyle{1\over3^n})}= -\sum_{n=1}^\infty (\textstyle{-1\over3})^n=(-1)\cdot{-1/3\over 1-(-1/3)} ={1/4}.$$

Now, the Cantor set is closed; and, since the Cantor set contains each $a_k$, it must contain the limit of $(a_k)$. Thus, the Cantor set contains the point $1/4$.

(This argument shouldn't seem so mysterious if you determine where $1/4$ lies in relation to the $a_k$; see the first paragraph of Michael Hardy's answer.)

David Mitra
  • 74,748
0

OP wanted "to know if $\frac14$ is in the Cantor set". Let us note $\mathbb K_3$ the Cantor set.

So, first of all, we have to tell what $\mathbb K_3$ is. As @Mickael Hardy said, "$\frac14$ is in the lower third. And $\frac14$ is in the upper third of the lower third. And in the lower third of that, and in the upper third of that, and so on." It is the well known geometrical description of Cantor set.

I propose then to formalize what Mickael said.

Let then $$C_0=D_0:=[0,1]$$ $$C_1=D_1=[0,\frac13]\cup[\frac23,1]$$

The idea I'm going to try to formalize is that if we enlarge three times from step to step, give or take, the non-integer part of $x$ must belong to $C_1$.

Let then$$D_2:=\{x\in [0,1]|9x-\left\lfloor 9x\right\rfloor\in C_1\}$$ $$C_2:=C_1\cap D_2$$ enter image description here and, by induction, $$\forall n\ge3, D_n:=\{x\in [0,1]|3^nx-\left\lfloor 3^nx\right\rfloor\in C_1\}$$ $$C_n:=C_{n-1}\cap D_n$$ Finally, classically, we define by$$\boxed{\mathbb K_3:=\bigcap_{n=1}^{+\infty}C_n}$$ It is then easy to use the base-three expansion of $x\in [0,1]$ $$x=\sum_{n=1}^{+\infty}\frac{a_n}{3^n}, a_i\in \left\{ 0,1,2 \right\}$$ (by not allowing degenerate expressions) to show that $$\boxed{x\in \mathbb K_3 \iff \forall n \in \mathbb N^*, a_n\in \{0,2\}}(*)$$


Then, concerning $\frac14=\sum_{n=1}^{+\infty}\frac{a_n}{3^n}, a_i\in \left\{ 0,1,2 \right\}$, (non degenerate development),

  1. $\frac14<\frac13$ and $\frac14=\frac{1\times 2}{4 \times 2}=\frac28$ and $\frac19<\frac18$. Then $\frac29<\frac14<\frac13$. So, $a_1=0$ and $a_2=2$

  2. If $n$ is even, $3^n \equiv 1 \mod 4$. Let us write $3^n=4q_n+1$(euclidean division). Then $\frac{3^n}{4}=q_n+\frac14=3^{n-1}a_1+3^{n-2}a_2+...+a_n+\frac{a_{n+1}}{3}+\frac{a_{n+2}}{3^2}+...$ Then $\frac14=\frac{a_{n+1}}{3}+\frac{a_{n+2}}{3^2}+...$ and $a_{n+1}=0 $ and $a_{n+2}=2$

  3. Then $$\boxed{\frac14=\frac03+\frac{2}{3^2}+\frac{0}{3^3}+\frac{2}{3^4}+...}$$ $$=(0,0:2:0:2:0:2:...)_{three}$$

  4. According to $(*)$, $$\boxed{\frac14 \in \mathbb K_3}.\square$$


Note that it was the search for the proof of the fact that $\frac14\in \mathbb K_3$, that gave me the idea of constructing $\mathbb K_3$ like this.

Stéphane Jaouen
  • 2,954
  • 5
  • 21