Show that f is uniformly continuous

  • Context: MHB 
  • Thread starter Thread starter evinda
  • Start date Start date
  • Tags Tags
    Continuous
Click For Summary

Discussion Overview

The discussion revolves around proving that a function \( f \) is uniformly continuous on a set \( A \) under certain conditions. Participants explore the implications of the definition of uniform continuity and how to construct sequences to demonstrate a contradiction if \( f \) is assumed not to be uniformly continuous.

Discussion Character

  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • Some participants propose that if \( x_n', x_n'' \in A \) and \( x_n' - x_n'' \to 0 \), then \( f(x_n'') - f(x_n'') \to 0 \) implies uniform continuity of \( f \) on \( A \).
  • Others argue that assuming \( f \) is not uniformly continuous leads to the existence of \( \epsilon > 0 \) such that for every \( \delta > 0 \), there are points \( y_n', y_n'' \in A \) with \( |y_n' - y_n''| < \delta \) and \( |f(y_n') - f(y_n'')| \geq \epsilon \).
  • Participants discuss the negation of the definition of uniform continuity and how it leads to the construction of sequences \( x'_n \) and \( x''_n \) that contradict the given property of \( f \).
  • There is a question about the choice of \( \delta = \frac{1}{n} \) and its relevance to obtaining sequences that converge to zero.
  • Some participants express uncertainty about whether to use sequences or other forms in their arguments.

Areas of Agreement / Disagreement

Participants generally agree on the approach to proving uniform continuity but express differing opinions on the specifics of the argument, particularly regarding the use of sequences and the implications of the definitions involved. The discussion remains unresolved as participants explore various angles without reaching a consensus.

Contextual Notes

Limitations include the dependence on the definitions of uniform continuity and the assumptions made about the sequences involved. The discussion does not resolve the mathematical steps or the implications of the arguments presented.

evinda
Gold Member
MHB
Messages
3,741
Reaction score
0
Hello! :cool:

I want to show that if $x_n',x_n'' \in A$ with $x_n'-x_n'' \to 0 \Rightarrow f(x_n'')-f(x_n'') \to 0$,then $f$ is uniformly continuous on $A$.

We suppose that $f$ is not uniformly continuous o $A$.
So, $ \exists \epsilon>0$ such that $\forall \delta>0$ and $ \forall y_n',y_n'' \in A$ with $|y_n'-y_n''|<\delta$ $\Rightarrow$ $|f(y_n')-f(y_n'')| \geq \epsilon$.
Is it right so far? And how can I continue? (Thinking)
 
Physics news on Phys.org
evinda said:
Hello! :cool:

I want to show that if $x_n',x_n'' \in A$ with $x_n'-x_n'' \to 0 \Rightarrow f(x_n'')-f(x_n'') \to 0$,then $f$ is uniformly continuous on $A$.

We suppose that $f$ is not uniformly continuous o $A$.
So, $ \exists \epsilon>0$ such that $\forall \delta>0$ and $ \forall y_n',y_n'' \in A$ with $|y_n'-y_n''|<\delta$ $\Rightarrow$ $|f(y_n')-f(y_n'')| \geq \epsilon$.
Is it right so far? And how can I continue? (Thinking)
When negating a statement containing quantifiers, each $\exists$ must be changed to $\forall$ and vice versa. The definition of uniform continuity states that $$ \forall \varepsilon>0\ \exists \delta>0\text{ such that } \forall x,x'\in A \text{ with } |x-x'|<\delta,\ |f(x) - f(x')| < \varepsilon.$$ The negation of that is that if $f$ is not uniformly continuous on $A$ then $$ \exists \varepsilon>0 \text{ such that }\forall \delta>0\ \exists x,x'\in A \text{ with } |x'-x''|<\delta,\ \text{ but }|f(x') - f(x'')| \geqslant \varepsilon.$$ Apply that statement with $\delta = 1/n$ ($n = 1,2,3,\ldots$). That will give you two sequences $x'_n$ and $x''_n$ that contradict the given property of $f$.
 
Opalg said:
When negating a statement containing quantifiers, each $\exists$ must be changed to $\forall$ and vice versa. The definition of uniform continuity states that $$ \forall \varepsilon>0\ \exists \delta>0\text{ such that } \forall x,x'\in A \text{ with } |x-x'|<\delta,\ |f(x) - f(x')| < \varepsilon.$$ The negation of that is that if $f$ is not uniformly continuous on $A$ then $$ \exists \varepsilon>0 \text{ such that }\forall \delta>0\ \exists x,x'\in A \text{ with } |x'-x''|<\delta,\ \text{ but }|f(x') - f(x'')| \geqslant \varepsilon.$$ Apply that statement with $\delta = 1/n$ ($n = 1,2,3,\ldots$). That will give you two sequences $x'_n$ and $x''_n$ that contradict the given property of $f$.

Why do we take $\delta=\frac{1}{n}$ ? Because of the fact that $\frac{1}{n} \to 0$ ? (Wondering)
 
evinda said:
Why do we take $\delta=\frac{1}{n}$ ? Because of the fact that $\frac{1}{n} \to 0$ ? (Wondering)
Yes. It is a convenient way to get two sequences $(x'_n)$ and $(x''_n)$ such that $x'_n-x''_n\to0.$
 
Opalg said:
Yes. It is a convenient way to get two sequences $(x'_n)$ and $(x''_n)$ such that $x'_n-x''_n\to0.$
So,we suppose that $f$ is not uniformly continuous on $A$.So,
$ \exists \varepsilon>0 \text{ such that }\forall \delta>0\ \exists y_n,y_n'\in A \text{ with } |y_n-y_n'|<\delta,\ \text{ but }|f(y_n) - f(y_n')| \geqslant \varepsilon.$

We pick $\delta=\frac{1}{n} \to 0$,and because of the fact that $y_n,y_n'\in A$ and $ |y_n-y_n'|<\delta \to 0$,we conclude that $f(y_n)-f(y_n') \to 0$,that is a contradiction,as $|f(y_n) - f(y_n')| \geqslant \varepsilon.$Right?? (Thinking)
 
evinda said:
So,we suppose that $f$ is not uniformly continuous on $A$.So,
$ \exists \varepsilon>0 \text{ such that }\forall \delta>0\ \exists y_n,y_n'\in A \text{ with } |y_n-y_n'|<\delta,\ \text{ but }|f(y_n) - f(y_n')| \geqslant \varepsilon.$

We pick $\delta=\frac{1}{n} \to 0$,and because of the fact that $y_n,y_n'\in A$ and $ |y_n-y_n'|<\delta \to 0$,we conclude that $f(y_n)-f(y_n') \to 0$,that is a contradiction,as $|f(y_n) - f(y_n')| \geqslant \varepsilon.$Right?? (Thinking)

Or shouldn't I have used sequences? (Thinking)
 
Last edited:
Or is it better like that? (Thinking)

We suppose that $f$ is not uniformly continuous on $A$.So,
$ \exists \varepsilon>0 \text{ such that }\forall \delta>0\ \exists y,y'\in A \text{ with } |y-y'|<\delta,\ \text{ but }|f(y) - f(y')| \geqslant \varepsilon. $

We take $\delta_n=\frac{1}{n}$ and then $\forall y_n,y_n' \in A$ and $|y_n-y_n'|<\delta \Rightarrow y_n-y_n' \to 0$, we conclude that $f(y_n)-f(y_n') \to 0 $,that is a contradiction.
 

Similar threads

  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 6 ·
Replies
6
Views
4K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
2
Views
2K