Why Does the Set {1/n : n ∈ ℕ} Have an Empty Interior?

In summary: That is a nice observation. From my understanding of discrete topology, it is a trivial topology where all subsets are open (and hence closed). Instead of open sets ∅, ℝ, we instead have open sets containing only one element. Correct?In summary, the conversation is about finding the interior point of a set, {1/n : n ∈ ℕ}, and determining that the answer is ∅ rather than (0,1). The reasoning behind this is that for any x in the set, there exists an irrational number that is also within the neighborhood of x, making it not an element of the set. This set is an example of a discrete set, where every point has a neighborhood intersecting it only at that point.
  • #1
zigzagdoom
27
0
Hi All,

A simple question but one for which I cannot seem to get the intuition.

1. Homework Statement


Find the interior point of {1/n : n ∈ ℕ}.

Homework Equations


N/A

The Attempt at a Solution



Let S = {1/n : n ∈ ℕ}, where S ⊆ℝ

x is an interior point if ∃N(x ; ε), N(x ; ε) ⊆ S.

My answer: IntS = (0,1)

But apparently the answer is ∅, which I do not seem to get.

Would it be that a small epsilon neighbourhood of x contains some element y such that y is not an element of S?

Any help is appreciated.

Edit:

I seemed to have figured it out. For every x for which we try to find the neighbourhood for, any ε > 0 we will have an interval containing irrational numbers which will not be an element of S.

Thanks
 
Last edited:
Physics news on Phys.org
  • #2
I assume that ##N(x;\varepsilon)## is the open ball of radius ##\varepsilon > 0## centered at ##x##?
zigzagdoom said:
My answer: IntS = (0,1)
Note that your answer could not possibly be correct, because for any set ##S##, in any topological space, it holds that ##\text{int}\,S \subseteq S##.
zigzagdoom said:
Would it be that a small epsilon neighbourhood of x contains some element y such that y is not an element of S?
Yes indeed. Try to make this precise. Take an arbitrary ##x = \frac{1}{n} \in S## and an arbitrary ##\varepsilon > 0##. Now indentify an element ##y \in \mathbb{R}## such that ##|x - y| < \varepsilon## but ##y \not\in S##. (You can write down an expression for such ##y## explicitly in terms of ##x## and ##\varepsilon##.)
 
  • Like
Likes zigzagdoom
  • #3
zigzagdoom said:
Edit:

I seemed to have figured it out. For every x for which we try to find the neighbourhood for, any ε > 0 we will have an interval containing irrational numbers which will not be an element of S.
Yes, well done!

Maybe it's also nice to know that a set ##A## in a topological space is called discrete when every point ##x \in A## has a neighborhood intersecting ##A## only in ##\{x\}##. So, ##S## is an example of a discrete set. However, there are sets (also in ##\mathbb{R}## with the usual metric) with empty interior that are not discrete. (And you know them, too, of course.)
 
  • Like
Likes zigzagdoom
  • #4
Krylov said:
I assume that ##N(x;\varepsilon)## is the open ball of radius ##\varepsilon > 0## centered at ##x##?

Note that your answer could not possibly be correct, because for any set ##S##, in any topological space, it holds that ##\text{int}\,S \subseteq S##.

Yes indeed. Try to make this precise. Take an arbitrary ##x = \frac{1}{n} \in S## and an arbitrary ##\varepsilon > 0##. Now indentify an element ##y \in \mathbb{R}## such that ##|x - y| < \varepsilon## but ##y \not\in S##. (You can write down an expression for such ##y## explicitly in terms of ##x## and ##\varepsilon##.)

Thanks a lot Krylov.

N(x; ε) is indeed the ball centre x, radius ε in this notation.

- (0,1) cannot be an interior point as 0 is actually a boundary point (i.e. ∀ N(0 ; ε), N(0 ; ε) ∧ ℝ \ S ≠ ∅)

- Let x ∈ S and x = 1/n. Now take any neighbourgood of x, N(x ; ε) = {y ∈ ℝ : |x - y| < ε}.
But for any n ∈ ℕ and any φ>0, there exists an irrational number i ∈ ℝ \ S, such that n - φ < i < n.
But then in any neighbourhood of x = 1/n, there exists a number 1/i. But 1/i ∉ S. Therefore there is no interior points of S, and IntS = ∅.
 
  • #5
Krylov said:
Yes, well done!

Maybe it's also nice to know that a set ##A## in a topological space is called discrete when every point ##x \in A## has a neighborhood intersecting ##A## only in ##\{x\}##. So, ##S## is an example of a discrete set. However, there are sets (also in ##\mathbb{R}## with the usual metric) with empty interior that are not discrete. (And you know them, too, of course.)
Thanks!
 

FAQ: Why Does the Set {1/n : n ∈ ℕ} Have an Empty Interior?

What is the "Interior Point of 1/n"?

The "Interior Point of 1/n" refers to a mathematical concept used in optimization problems. It is a point within the feasible region of a convex optimization problem that satisfies all of the constraints and minimizes the objective function, where n is the number of variables in the problem.

How is the "Interior Point of 1/n" calculated?

The "Interior Point of 1/n" is typically calculated using interior point methods, which involve iteratively moving towards the optimal point within the feasible region. These methods use a barrier function to ensure that the point remains within the feasible region and a centering function to minimize the objective function.

What is the significance of the "Interior Point of 1/n"?

The "Interior Point of 1/n" is significant because it represents the optimal solution to a convex optimization problem. This optimal solution can be used to make informed decisions and improve systems in various fields, including engineering, economics, and data science.

Can the "Interior Point of 1/n" be used in non-convex optimization problems?

No, the "Interior Point of 1/n" method is only applicable to convex optimization problems. Non-convex problems have multiple local minima, making it difficult to determine the optimal solution using this method.

Are there any limitations to using the "Interior Point of 1/n"?

While the "Interior Point of 1/n" method is effective for convex optimization problems, it can be computationally expensive and may not always find the global optimal solution. Additionally, it may not work for problems with non-smooth or non-continuous objective functions.

Back
Top