2
$\begingroup$

Let $f: X \to Y$ be a function between metric spaces $X, \phi$ and $Y, d$ such that $f^{-1}(U)$ is open in $X$ for every subset $U$ open in $Y$. Prove that if $C$ is a compact subset of $X$, then $f(C)$ is a compact subset of $Y$.

My attempt: Let $C$ be a compact subset in $X, \phi$. Let $\{V_{\alpha}\}$ be a collection of open sets such that $C^{cpct} \subset \bigcup_{\alpha} V_{\alpha}$. Then there exist $\alpha = 1, 2, ..., n$ such that $C^{cpct} \subset \bigcup_{\alpha} V_{\alpha_n} : n \in \mathbb{N}$.

I want to show this subcover is open and thus it has an open image, but I'm not sure as if that's correct because only the converse is true. Is it true?

I need way better notation. Also, how to continue - any suggestions?

$\endgroup$
  • $\begingroup$ By definition, f is compact, since the inverse image of every open subset in $Y$ is open in $X$. Then $f(C)$ is the continuous image of compact, so it is compact. $\endgroup$ – user99680 Nov 6 '13 at 5:23
  • $\begingroup$ @user99680 I think you mean $f$ is continuous. I also think that this problem is asking for the proof that the continuous image of a compact set is compact. $\endgroup$ – manthanomen Nov 6 '13 at 5:25
  • $\begingroup$ @manthanomen no (prove that this function is continuous and I shall award you a bounty of +50.) $\endgroup$ – Don Larynx Nov 6 '13 at 5:26
  • $\begingroup$ @DonLarynx Care to elaborate? It's continuous because the inverse images of open sets are open. $\endgroup$ – manthanomen Nov 6 '13 at 5:29
  • $\begingroup$ @Don Larynx: what is your definition of continuity? One I'm familiar with is that the inverse image of every open set is open. $\endgroup$ – user99680 Nov 6 '13 at 5:39
3
$\begingroup$

You wish to show the image of a compact set is compact under a continuous mapping. This can be done by taking finite sub-covers. Let $\{\mathcal{O}_\alpha\}_{\alpha\in A}$ be an open cover of $f(C)$. Then $\{\mathcal T_\alpha\}_{\alpha \in A}$ is an open covering of $C$, where $T_\alpha \equiv f^{-1}(\mathcal{O}_\alpha)$ for each $\alpha \in A$.

Now choose $\alpha_1 \ldots \alpha_n$ for some $n \in \mathbb{N}$ such that $C \subset \cup_{i=1}^n T_{\alpha_i}$. This can be done since $C$ is compact.

I will leave it to you to show that $\cup_1^n f(T_{\alpha_i}) = \cup_1^n \mathcal{O}_{\alpha_i} \supset f(C)$.

Since we have taken an arbitrary open cover of $f(C)$ and found a finite subcover, then $f(C)$ must be compact.

Since we have chosen an arbitrary compact set $C$ and shown $f(C)$ is compact, then the image of any compact set is compact under $f$.

And finally, since $f$ is an arbitrary continuous function, we have shown that continuous functions map compact sets to compact sets.

$\endgroup$
  • $\begingroup$ Thank you so so much. I was able to use the fact $f(\cup\scr{C}) = \cup f(\scr{C})$! $\endgroup$ – Don Larynx Nov 12 '13 at 16:48
1
$\begingroup$

The assumption that $f^{-1}(U)$ is open in $X$ if $U$ is open in $Y$ tells us that the function is continuous. But it doesn't tell us that the images of open sets are open (i.e., $f$ is not necessarily an open map). But all you need to know for the proof is that inverse images of open sets are open.

Try to prove the contrapositive statement:

Assume $f(C$) is not compact. So there exists an irreducible open cover $\displaystyle\bigcup_\alpha U_\alpha$ of $f(C)$. Now consider $\displaystyle\bigcup_\alpha f^{-1}(U_\alpha)$. With a little work, you'll see that this is an irreducible cover of $C$, and so $C$ is not compact.

$\endgroup$
0
$\begingroup$

We can see that map $f$ is a continuous , for a continuous i have some equivalent conditions in metric space:

$a)$ $f$ is a continuous map

$b)$ The image of an open set is an open set

$c)$ The image of an close set is an close set .

So $f$ is continuous , i will show that if $C$ is a compact set then $f(C)$ is a compact set , I choose an any infinity sequence $f(x_1),f(x_2),..f(x_n),,,$ in $f(C)$ , because $C$ is compact so i can choose $(x_{k_n})$ of $(x_n)$ that converging and $f$ is continuous then $f(x_{k_n})$ is converging . Finally i have $f(C)$ is compact set

$\endgroup$
  • $\begingroup$ Pham, please check the dates on questions before you provide answers to them, especially if there are already other answers available. This question is nearly three years old, and answering old questions moves them onto the main page, which bumps newer, unanswered questions into areas of the site with less visibility. $\endgroup$ – Eric Stucky Oct 4 '16 at 15:47
  • $\begingroup$ I sure that about my answer $\endgroup$ – Gankedbymom Oct 4 '16 at 16:01
  • $\begingroup$ Please do not bump old questions which already have accepted answers. Your answer is correct; this is not relevant to my point. $\endgroup$ – Eric Stucky Oct 4 '16 at 16:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.