Equicontinuity and uniform convergence

Click For Summary

Homework Help Overview

The problem involves demonstrating that a sequence of functions from C(X, Y) converging uniformly implies that the collection is equicontinuous. The context centers around concepts of uniform convergence and equicontinuity in the setting of metric spaces.

Discussion Character

  • Conceptual clarification, Assumption checking

Approaches and Questions Raised

  • Participants discuss the necessity of compactness in the space X, with some questioning whether the proof holds without it. There are attempts to clarify the implications of uniform continuity and its relation to the proof.

Discussion Status

The discussion is active, with differing views on the role of compactness and uniform continuity. Some participants express confidence in the proof's correctness, while others raise concerns about the assumptions involved.

Contextual Notes

There is a mention of a potential error in a definition from Munkres, which may affect the understanding of pointwise boundedness in the context of the problem.

radou
Homework Helper
Messages
3,149
Reaction score
8

Homework Statement



This is a nice one, if it's correct.

Show that if (fn) is a sequence of elements of C(X, Y) (where Y is a metric space) which converges uniformly, then the collection {fn} is equicontinuous.

The Attempt at a Solution



Let ε > 0 be given and let x0 be a point of X. The proof is based on the following:

d(fn(x0), fn(x)) <= d(fn(x0), f(x0)) + d(f(x0), f(x)) + d(f(x), fn(x)).

Since fn converges uniformly, we can find a positive integer N such that, for all x in X, and for n >= N, we have d(f(x), fn(x)) < ε/3. Since f is continuous, for ε/3 choose a neighborhood U of x0 such that whenever x is in U, we have d(f(x0), f(x)) < ε/3. Now the upper inequality becomes d(fn(x0), fn(x)) < ε, for all x in X (and hence for all x in U), and for all n >= N, i.e. for all but finitely members of the collection {fn}. Now, for every of the remaining members of {fn], use continuity of fn at x0 to find neighborhoods of x0 such that d(fn(x0), fn(x)) < ε. Now, intersect this finite number of neighborhoods of x0 with the neighborhood U, call this intersection U'. Clearly, for any x in U, we have d(fn(x0), fn(x)) < ε, for all elements of {fn}.
 
Physics news on Phys.org
Is X compact here? Because if it is not, i don't think the statement is even true. The problem is that if X is not compact than f may or may not be uniformly continuous, which is crucial in this case. For, if f is not uniformly continuous you won't be able to find an epsilon that will work for all N and all x.

Other than this, your proof seems correct.
 
No, Y is not assumed to be compact.
 
The proof is correct. I don't think you need compactness or uniform continuity here...
 
micromass said:
The proof is correct. I don't think you need compactness or uniform continuity here...

Thanks!

Btw, it seems I found the first error in Munkres, after 278 pages!

Definition. If (Y, d) is a metric space, a subset F of C(X, Y) is said to be pointwise bounded under d if for each x in X, the subset Fa = {f(a) : f is in F} of Y is bounded under d.

I assume he meant Fx = {f(x) : f is in F}, right? I mean, it's not a big mistake (if!) but I want to make sure, since I need this definition in another proof...
 
Haha! You're correct, Munkres indeed made a mistake here :smile:
 
micromass said:
Haha! You're correct, Munkres indeed made a mistake here :smile:

OK! Now off to proving Arzela's theorem... :biggrin:
 

Similar threads

  • · Replies 3 ·
Replies
3
Views
2K
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
Replies
8
Views
3K
Replies
11
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K