What is the Limit in a Function Space?

Click For Summary

Homework Help Overview

The discussion revolves around a limit in the context of metric spaces, specifically focusing on the uniform convergence of functions defined on these spaces. The original poster presents a problem involving the continuity of functions and the behavior of sequences within metric spaces.

Discussion Character

  • Conceptual clarification, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • Participants explore the implications of uniform convergence and continuity, questioning the application of the triangle inequality in the context of limits. There are attempts to establish relationships between the limits of function sequences and their continuity properties.

Discussion Status

The discussion is active, with participants providing insights and corrections regarding the original poster's reasoning. Some guidance has been offered regarding the proper use of inequalities and continuity, indicating a productive exploration of the problem.

Contextual Notes

There is an emphasis on the continuity of the functions involved and the convergence of sequences within the metric spaces. Participants are also considering the implications of the definitions and properties of limits in this context.

jdcasey9
Messages
25
Reaction score
0

Homework Statement



Let (X,d) and (Y,p) be metric spaces, and let f, fn: X -> Y with fn -> f uniformly on X. If each fn is continuous at xcX, and if xn -> x in X, prove that lim n-> infinity fn(xn) = f(x).

Homework Equations



llxll inf (the infinity norm of x) = max (lx1l,...,lxnl)

The Attempt at a Solution



fn and f are continuous, so f(x) is defined.

lfn(xn) - f(x)l <= llfn-fll inf (the infinity norm of fn-f) and llfn-fll inf -> 0 as n->infinity so lfn(xn)-f(x)l -> 0 as n -> infinity and lim fn(xn) = f(x) as n->infinity.

Seems very easy, so I probably am missing something.
 
Physics news on Phys.org
Yeah, there's a small thing you've missed and I have the feeling you should have used the triangle inequality somewhere.. You say that

|f_n(x_n)-f(x)|\leq \|f_n-f\|_\infty

But this is'nt true. It is of course true that

|f_n(x)-f(x)|\leq \|f_n-f\|_\infty

So you just need to use the triangle inequality somehow...
 
Alright,

lfn(xn) - f(x)l <= lfn(xn) - fn(x)l + lfn(x) - f(x)l <= lfn(xn) - fn(x)l + llfn - fll inf -> lfn(xn) - fn(x)) + 0 as n->inf.

So, now we need to show that lfn(xn) - fn(x)l -> 0.

Can we do this the same way?

lfn(xn) - fn(x)l <= lfn(xn) - f(xn)l + lf(xn) - fn(x)l -> lf(xn) - fn(x)l <= lf(xn) -f(x)l + lfn(x) - f(x)l -> lf(xn) - f(x)l -> 0 as n->inf

Can we say that lfn(xn) - f(xn)l -> llfn - fll inf norm? and that lf(xn) - f(x)l -> 0 as n-> inf? (without any further proof)
 
To prove that |f_n(x_n)-f_n(x)|\rightarrow 0.

We know that x_n\rightarrow x. Continuity of the f_n yields that f_n(x_n)\rightarrow f_n(x).
 
Oh, ok, thanks I appreciate your help.
 

Similar threads

  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 10 ·
Replies
10
Views
1K
  • · Replies 13 ·
Replies
13
Views
2K
Replies
8
Views
3K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
Replies
11
Views
3K
  • · Replies 11 ·
Replies
11
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
Replies
2
Views
1K