Does Compactness Depend on Metric Choice in Metric Spaces?

  • Thread starter JG89
  • Start date
  • Tags
    Metric
In summary, compactness of a subset A in a metric space X does depend on the given metric. For example, if there are two possible metrics, d(x,y) and p(x,y), A is compact with respect to d(x,y) if and only if it is compact with respect to p(x,y). This can be seen through examples such as the discrete metric, where compact sets are exactly the finite sets, and the usual metric on real numbers, where the interval [0,1] is compact but not with the discrete metric. Additionally, a subset A in a metric space X can be compact with the usual metric but not with the discrete metric, as seen in the example of the sequence {0, 1,
  • #1
JG89
728
1
Suppose that A is a subset of a metric space X. Does compactness of A depend on which metric is given to X? For example, if d(x,y) and p(x,y) are two possible metrics for X, is A compact with respect to the metric d(x,y) if and only if A is compact with respect to the metric p(x,y)?
 
Physics news on Phys.org
  • #2
I am no expert on this forum but I believe that if there is an open cover of A in one metric, the other will follow
 
  • #3
Yes, "compactness" does depend strongly on the metric. A simple example is the metric d(x,y)= 1 if [itex]x\ne y[/itex], 0 otherwise. That is called the "discrete metric" because using that metric on any set gives the "discrete topology" in which all sets are open. To see that, recognize that the "neighborhood of point p with radius 1/2" is just the singleton set {p} itself- since the distance from p to any other point is 1, the only point, x, for which it is true that d(p, x)< 1/2 is p itself. Every point, in every set, is an interior point because that singleton set is a subset neighborhood contained in the set.

If X is an infinite set, with the discrete metric, then the compact sets are exactly the finite sets. That is because we could always use the individual singleton sets, {a} where a is any member of A, as our open cover. Since every point of A is in only one of those, we cannot remove any of them, much less reduce to a finite cover.


So, for example, take X to be the set of all real numbers.

With the "usual metric", d(x,y)= |x- y|, the interval [0, 1] is compact because it is closed and bounded. With the discrete metric, it is not compact because it is not finite.

Another example: X is the set of real numbers formed by the sequence {0, 1, 1/2, 1/3, ..., 1/n, ...}, again with the usual metric on the real numbers. Let [itex]\{U_n\}[/itex] be any open cover. Since 0 is in A, there exist some [itex]U_0[/itex] which contains 0. Since [itex]U_0[/itex] is open, 0 is an interior point- there exist [itex]\delta[/itex] such that [itex]\{x | |x|< \delta\}[/itex] is a subset of [itex]U_0[/itex]. But the sequence 1, 1/2, 1/3, ... converges to 0. There exist some N such that if n> N, [itex]|1/n|< \delta[/itex] and so all 1/n, for n> N, is in [itex]U_o[/itex]. Pick a single [itex]U_n[/itex] that contains 1/n for n< N. That is a finite collection and it, together with [itex]U_0[/itex], makes a finite subcollection that covers A. A is compact.

But that same set of real numbers, with the discrete metric, is infinite and so is not compact.
 
Last edited by a moderator:
  • #4
Great reply Halls. That helped a lot!
 

FAQ: Does Compactness Depend on Metric Choice in Metric Spaces?

What is compactness in metric spaces?

Compactness in metric spaces refers to a property of a set that describes how well it can be covered by a finite number of smaller sets. In other words, it measures the degree to which a set is "small" or "finite" in a given metric space.

How is compactness different from connectedness?

While compactness measures the "size" of a set, connectedness measures the "integrity" or "connectedness" of a set. A set can be compact without being connected, and vice versa.

What is the compactness theorem?

The compactness theorem, also known as the Heine-Borel theorem, states that in a metric space, a set is compact if and only if it is closed and bounded. This theorem is a fundamental result in analysis and is often used to prove other theorems.

What are some examples of compact sets?

In Euclidean spaces, closed and bounded intervals are compact, as well as closed and bounded circles or spheres. Other examples include finite sets and Cantor sets. In general, any set that can be covered by a finite number of smaller sets is considered compact.

Why is compactness important in mathematics?

Compactness is a key concept in many areas of mathematics, including analysis, topology, and geometry. It allows us to establish important properties of sets and functions, and it is often used to prove theorems and solve problems. In particular, compactness is essential in the study of continuity, convergence, and completeness in metric spaces.

Similar threads

Replies
5
Views
739
Replies
5
Views
2K
Replies
2
Views
3K
Replies
32
Views
2K
Replies
3
Views
1K
Replies
3
Views
1K
Replies
2
Views
2K
Replies
1
Views
879
Replies
2
Views
711
Back
Top