Prove that $f : A\subset \mathbb{R} \to \mathbb{R}$ is uniformly continuous in $A$ $ \iff$ for all

the sequences $(x_n), (y_n)\in A$ such that $$\lim_{n\to +\infty} ( x_n - y_n )= 0$$


$$\lim_{n\to +\infty} ( f(x_n) - f(y_n)) = 0.$$


Let's start with the first to demonstrate the implication:


in this case the hp are:

$ f $ uniformly continuous and $ (x_n), (y_n) $ sequences of numbers from the set $ A $ such that: $$\lim_{n\to+\infty}(x_n - y_n) = 0.$$

and we must show that: $$ \lim_{n\to +\infty} ( f(x_n) - f(y_n)) = 0.$$ Well, for the uniform continuity we have $$\forall\varepsilon> 0, \exists\delta> 0 \, \: \, \ \forall x, y \in A \ | x-y | <\delta \Rightarrow | f (x)-f (y) | <\varepsilon,$$

and by definition of limit: $$ \forall \varepsilon> 0, \exists \nu> 0 \, \, \, \: \, \, \, \ | (x_n-y_n) - 0 | = | x_n-y_n | <\varepsilon, \quad n> \nu $$

but then the fact that, thanks to the uniform continuity we have, $$ | x-y | <\delta \Rightarrow | f (x)-f (y) | <\varepsilon, $$

we have that $$ | f (x_n) - f (y_n) | <\varepsilon, \forall n> \nu $$

which is equivalent to saying

$$ \lim_ {n \to + \infty} (f (x_n) - f (y_n)) = 0. $$ $ "{\Leftarrow}" $

in this case hp are: $$ (1) \lim_{n\to +\infty} ( f(x_n) - f(y_n)) = 0 $$ $$ (2) \lim_{n\to+\infty}(x_n - y_n) = 0.$$ and we must show that $f$ is uniformly continuous: but if (1) and (2) are verified, $f$ is necessarily uniformly continuous . QED


Your forward implication is a bit off. In particular, you want $\delta$, not $\epsilon$, in the fourth displayed equation of your proof.

You also need to explicitly show that for any $\epsilon>0$, there is a $N$ (your $\nu$) that "works". You also need to explicitly fix your $\delta$.

You fix a value of $\epsilon$ first, and then state (and prove) that the appropriate $\delta$ and $N$ exist. Then you verify that the choices "work".

So, you could argue as follows:

Let $\epsilon>0$. Then, since $f$ is uniformly continuous on $A$, there is a $\delta>0$ (which we are now specifying) such that $$\tag{1}|x-y|<\delta \Rightarrow |f(x)-f(y)|<\epsilon \quad\text{ for all }x,y \text{ in } A .$$

Now, choose $N$ (as we may) so that $$\tag{2}|x_n-y_n|<\delta\quad\text{ for all }n>N.$$

We have found our $N$. Now we need to show that it actually works:

Then, if $n>N$, we have, by (2), that $|x_n-y_n|<\delta$. Then, from (1), we have $|f(x_n)-f(y_n)|<\epsilon$.

So we have shown that for any $\epsilon>0$, there is an $N$ so that $n>N$ implies $|f(x_n)-f(y_n)|<\epsilon$. Thus $(f(x_n)-f(y_n))\rightarrow 0$, as desired.

For the reverse implication, it seems that you are assuming your result...

Some hints:

Prove the contrapositive statement instead.

To do this, you could do the following:

Suppose that $f$ is not uniformly continuous on $A$. Then there would be an $\epsilon>0$ such that for any positive integer $N$, you can find points $x_n$ and $y_n$ with $|x_n-y_n|<{1\over N}$ but $|f(x_n) -f(y_n)|>\epsilon$.


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.