Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top


It is a standard and important fact in basic calculus/real analysis that a continuous function on a compact metric space is in fact uniformly continuous. That is, suppose $(X,d)$ is a compact metric space and $f\colon X \to\mathbb R$ is such that for every $x\in X$ and $\varepsilon>0$ there exists $\delta>0$ such that $d(x,y)<\delta$ implies $|f(x)-f(y)|<\varepsilon$. Then in fact, such a $\delta$ can be chosen independently of $x^.


Does a similar statement hold regarding semi-continuous functions? For concreteness, let's consider upper semi-continuous functions, so suppose $(X,d)$ is compact and $f\colon X \to\mathbb R$ has the property that for every $x\in X$ and $\varepsilon >0$ there exists $\delta >0$ such that $d(x,y)<\delta$ implies $f(y) < f(x)+\varepsilon$. (Note the asymmetry of $x$ and $y$ in this definition.) Then is it true that $\delta=\delta(\varepsilon)$ can be chosen independently of $x$?


Given $\delta, \epsilon > 0$, consider the set $$ X_\delta^\epsilon := \lbrace x\in X \mid f(y) < f(x) + \epsilon \text{ for every } y\in B(x,\delta) \rbrace. $$ Then $f$ is upper semi-continuous if and only if $\displaystyle\bigcup_{\delta>0} X_\delta^\epsilon = X$ for every $\epsilon > 0$, and $f$ is uniformly upper semi-continuous if and only if this union stabilises -- that is, if for every $\epsilon > 0$ there exists $\delta>0$ such that $X_\delta^\epsilon = X$.

share|cite|improve this question
up vote 4 down vote accepted

$f(x)=0 \ (x\le0)$, $f(x)=-1/x \ (x\gt0)$ is upper semi-continuous on $[-1;1]$ — but not uniformly.

share|cite|improve this answer
Of course... thanks for the example. Do you know what happens if we require the function f to be bounded below? (In the context that motivated the question, I'm considering such functions.) – Vaughn Climenhaga Aug 9 '10 at 14:09
Turns out boundedness doesn't help (see my answer). In retrospect this all seems quite obvious... maybe this is why I shouldn't post questions in the wee hours of the morning. – Vaughn Climenhaga Aug 9 '10 at 15:10

A little further thought reveals the following: uniform semi-continuity implies uniform continuity. Thus the answer to my question is a resounding "no", since any function that is upper semi-continuous but not continuous cannot be uniformly upper semi-continuous.

Proof. Let $f$ be uniformly upper semi-continuous. Then for every $\epsilon>0$ there exists $\delta>0$ such that for every $x\in X$, we have $f(y) < f(x) + \epsilon$ whenever $y\in B(x,\delta)$. However, since this statement holds for every $x$, it also holds with $x$ and $y$ reversed; in the language of the original post, both $x$ and $y$ are contained in the set $X_\delta^\epsilon = X$. Since $y$ is in this set and $x\in B(y,\delta)$, we also have $f(x) < f(y) + \epsilon$, and thus $|f(x) - f(y)| < \epsilon$. But this is just the definition of uniform continuity.

share|cite|improve this answer
ah! (I should have thought about this instead of constructing artificial counter-examples) – Grigory M Aug 9 '10 at 15:22

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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