Proving Convergence in C[0,1] with Integral Norm

In summary: Consider the definition of the infinity norm in a function space:$$||f||_\infty = \sup|f(x)|$$This means that for any x in the interval [0,1], |f(x)| will be less than or equal to the infinity norm. Therefore, we can say that:$$\int_0^1|f_n(x)-f(x)|\, dx \le \int_0^1||f_n(x)-f(x)||_\infty\, dx$$Since we already know that for any ε > 0, there is an N0 such that, if n > N0, then ||fn - f||∞ < 0, we
  • #1
bugatti79
794
1

Homework Statement



Show that a sequence ##f_n \to f \in C[0,1]## with the sup norm ##|| ||_\infty##, then ##f_n \to f \in C[0,1]## with the integral norm.

The Attempt at a Solution



given ##\epsilon > 0 \exists n_0 \in N## s.t

##||(fn-f) (x)|| < \epsilon \forall n > n_0## with ## x \in [a,b]##

ie ##\forall \epsilon > 0 \forall n_0 \in N## s.t

##sup |(f_n-f)(x)|=sup|f_n(x)-f(x)|## with ##x \in C[0,1]## and ##n \in \mathbb{N} \implies f_n \to f \in C[0,1]##

if this is correct, do I attempt the same for the integral norm?
 
Physics news on Phys.org
  • #2
It looks to me like there are two words missing in your problem statement - "if" and "then".
bugatti79 said:

Homework Statement



Show that a sequence ##f_n \to f \in C[0,1]## with the sup norm ##|| ||_\infty##, then ##f_n \to f \in C[0,1]## with the integral norm.

The Attempt at a Solution



given ##\epsilon > 0 \exists n_0 \in N## s.t

##||(fn-f) (x)|| < \epsilon \forall n > n_0## with ## x \in [a,b]##

ie ##\forall \epsilon > 0 \forall n_0 \in N## s.t

##sup |(f_n-f)(x)|=sup|f_n(x)-f(x)|## with ##x \in C[0,1]## and ##n \in \mathbb{N} \implies f_n \to f \in C[0,1]##

if this is correct, do I attempt the same for the integral norm?
I believe what the problem is saying is this:
IF fn → f using the || || norm, THEN fn → f using the integral norm.

If my surmise is correct, what you have done is prove the hypothesis, which is done only in if and only if proofs.
 
  • #3
Mark44 said:
It looks to me like there are two words missing in your problem statement - "if" and "then".
I believe what the problem is saying is this:
IF fn → f using the || || norm, THEN fn → f using the integral norm.

If my surmise is correct, what you have done is prove the hypothesis, which is done only in if and only if proofs.

Yes, your surmise is correct. So what you recommend I do? I need to show some link between between the sup and integral norm, right? Not sure how to do that...thanks
 
  • #4
Given that for any ε > 0, there is an N0 such that, if n > N0, then ||fn - f|| < 0,

show that for any ε > 0, there is an M0 such that, if n > M0,
[tex]\int_0^1 f~f_n~dt < \epsilon[/tex]

You'll need to use the definition of the infinity norm to establish the second inequality.
 
  • #5
Mark44 said:
Given that for any ε > 0, there is an N0 such that, if n > N0, then ||fn - f|| < 0,

show that for any ε > 0, there is an M0 such that, if n > M0,
[tex]\int_0^1 f~f_n~dt < \epsilon[/tex]

You'll need to use the definition of the infinity norm to establish the second inequality.

The definition of an infinity norm in a function space is

##||f||_\infty= sup | f(x)|## where ##x \in C[a,b]##

Not sure how to proceed further especially when I see a product of ##f*f_n##...?
 
  • #6
How is the integral norm defined? I just took a guess at how it would look.
 
  • #7
Mark44 said:
Given that for any ε > 0, there is an N0 such that, if n > N0, then ||fn - f|| < 0,

He means ##< \epsilon##

show that for any ε > 0, there is an M0 such that, if n > M0,
[tex]\int_0^1 f~f_n~dt < \epsilon[/tex]

You'll need to use the definition of the infinity norm to establish the second inequality.

bugatti79 said:
The definition of an infinity norm in a function space is

##||f||_\infty= sup | f(x)|## where ##x \in C[a,b]##

Not sure how to proceed further especially when I see a product of ##f*f_n##...?

Hopefully you could figure out he meant$$
\int_0^1|f_n(x)-f(x)|\, dx <\epsilon$$
 
  • #8
Yes, I definitely meant ϵ, not 0.

And I misremembered the integral norm. I think I confused it with one of the inner products.
 
  • #9
LCKurtz said:
He means ##< \epsilon##





Hopefully you could figure out he meant$$
\int_0^1|f_n(x)-f(x)|\, dx <\epsilon$$

I can't see the link. All that's going through my head is for the integral norm

##\displaystyle \int_0^1 |f_n(x)-f(x)| dx = | \frac{|f_n(x)-f(x)|^{n+1}}{n+1}|_0^1## which would evaluate to some value < ε if the assumption ##f_n(x) \to f(x)## is true

and for the infinity norm

## sup|f_n(x)-f(x)| = max{|f_n(x)-f(x)|}## for ##x \in C[0,1]## which would evaluate to some value < ε if the assumption ##f_n(x) \to f(x)## is true also...?
 
  • #10
bugatti79 said:
I can't see the link. All that's going through my head is for the integral norm

##\displaystyle \int_0^1 |f_n(x)-f(x)| dx = | \frac{|f_n(x)-f(x)|^{n+1}}{n+1}|_0^1##

Oh my. After seeing that I think the best suggestion I can make is for you to schedule a meeting with your instructor. I don't see how to continue helping you with this problem without working it for you. Do you have the appropriate prerequisites for the course these problems are coming from?
 
  • #11
LCKurtz said:
Oh my.
Exactly my thought.

IIRC, Bugatti79 is studying this stuff on his own.
 
  • #12
bugatti79 said:
##\displaystyle \int_0^1 |f_n(x)-f(x)| dx = | \frac{|f_n(x)-f(x)|^{n+1}}{n+1}|_0^1##
I know well this in not correct mathematically, I shouldn't have wrote that. I was just trying to indicate that if we integrate this function between the limits 0 and 1 we should get some value less than some (ε>0) that we have chosen. Of course, that doesn't make me progress any further.
 
  • #13
Mark is this was you aim at:

[tex] \int_0^1 |f_n(x)-f(x)| dx \leq \int_0^1 ||f_n(x)-f(x)||_{\infty} dx =\epsilon\ \forall n \geq N_0 [/tex]
 
  • #14
dirk_mec1 said:
Mark is this was you aim at:

[tex] \int_0^1 |f_n(x)-f(x)| dx \leq \int_0^1 ||f_n(x)-f(x)||_{\infty} dx =\epsilon\ \forall n \geq N_0 [/tex]

Should that be "##< \epsilon##" not "##= \epsilon##"?
 
  • #15
Yes.
 
  • #16
I have just put the information together for completeness, I am still wondering how to finish it.?

Show that a sequence fn→f∈C[0,1] with the sup norm ||||∞, then fn→f∈C[0,1] with the integral norm.

Given that for any ε > 0, there is an N0 such that, if n > N0, then ||fn - f||∞ < 0,

show that for any ε > 0, there is an M0 such that, if n > M0 then

$$\int_0^1|f_n(x)-f(x)|\, dx <\epsilon$$..?
 
  • #17
The relevant inequality for this problem is [itex]|f_n(x)-f(x)| \leq ||f_n-f||_{\infty}[/itex] for all [itex]x \in [0,1][/itex]. But honestly, based on your threads in the homework help forums, you really lack the mathematical maturity to be dealing with convergence in function spaces. It would be worthwhile for you to go back and review basic convergence in [itex]\mathbb{R}[/itex].
 
  • #18
jgens said:
The relevant inequality for this problem is [itex]|f_n(x)-f(x)| \leq ||f_n-f||_{\infty}[/itex] for all [itex]x \in [0,1][/itex]. But honestly, based on your threads in the homework help forums, you really lack the mathematical maturity to be dealing with convergence in function spaces. It would be worthwhile for you to go back and review basic convergence in [itex]\mathbb{R}[/itex].

I have looked through my notes and through Erwin Kreyszig's 'Introductory Functional Analysis'. Can you be more specific what you mean by basic convergence in R, so I can check the book.

Thanks
 
  • #19
bugatti79 said:
I have looked through my notes and through Erwin Kreyszig's 'Introductory Functional Analysis'.

Unless you have some secret talent you are not letting on to here, then that book is much too difficult for you.

Can you be more specific what you mean by basic convergence in R, so I can check the book.

I should have been more precise. I meant convergence in the p-norms on [itex]\mathbb{R}^n[/itex].
 

FAQ: Proving Convergence in C[0,1] with Integral Norm

What is function sequence convergence?

Function sequence convergence is a mathematical concept that refers to the idea of a sequence of functions approaching a particular value or function as the input variable approaches a specific value or infinity.

How is function sequence convergence different from pointwise convergence?

Pointwise convergence refers to the idea that each individual point in a sequence of functions approaches a specific value as the input variable approaches a particular value or infinity. Function sequence convergence, on the other hand, refers to the entire sequence of functions approaching a value or function as a whole.

What are the most common methods for proving function sequence convergence?

The most common methods for proving function sequence convergence include the use of the epsilon-delta definition, the monotone convergence theorem, and the Cauchy convergence criterion.

What are some real-world applications of function sequence convergence?

Function sequence convergence has numerous real-world applications in fields such as physics, engineering, and economics. For example, it can be used to model the behavior of systems that change over time, such as the decay of radioactive materials or the growth of populations.

Are there any limitations or conditions for function sequence convergence?

Yes, there are certain conditions that must be met for function sequence convergence to be true. These include the need for the functions in the sequence to be defined on the same domain and for the sequence to be uniformly convergent. Additionally, some functions may not converge at all, or may only converge under certain conditions.

Similar threads

Replies
4
Views
994
Replies
7
Views
2K
Replies
8
Views
1K
Replies
5
Views
2K
Replies
7
Views
2K
Replies
6
Views
1K
Replies
18
Views
2K
Replies
2
Views
1K
Back
Top