Is the subset G totally bounded? - Proving or disproving using relevant theorems

  • Thread starter Eulogy
  • Start date
  • Tags
    Bounded
In summary: You can define a function f so that its integral is 1/2, but it is only nonzero on [1/2,1]. Same thing for f2 but it is only nonzero on [1/4,1/2], put f3 on [1/8,1/4] etc.
  • #1
Eulogy
8
0

Homework Statement


Let G = { f [itex]\in[/itex] C[0,1] : [itex]^{0}_{1}[/itex][itex]\int[/itex]|f(x)|dx [itex]\leq[/itex] 1 }
Endowed with the metric d(f,h) = [itex]^{0}_{1}[/itex][itex]\int[/itex]|f(x)-h(x)|dx. Is G totally bounded? Prove or provide counterexample

2. Relevant Theorems

Arzela-Ascoli Theorem, Theorems relating to compactness, equicontinuity etc and

Let M be a compact metric space. A subset of
C(M) is totally bounded iff it is bounded and equicontinuous. (in terms of the sup metric)

The Attempt at a Solution


I'm not to sure whether G is totally bounded or not to begin with. If I was working in the sup metric the subset is not totally bounded as I can find a family of functions which are not equicontinuous in G. However I'm not sure how or if this translates to the integral metric defined above. Any help would be much appreciated!
 
Physics news on Phys.org
  • #2
Try to find a sequence of functions [itex]f_i(x)[/itex] such that [itex]d(f_i,f_j)=1[/itex] for all [itex]i \ne j[/itex]. Step functions will do. What would that tell you about total boundedness?
 
  • #3
Such a sequence could not be totally bounded since if I took a finite cover of the space with balls radius say half then each function in the sequence must be in a separate ball and as the sequence is infinite this leads to a contradiction. But I can't think of a sequence functions that are continuous on [0,1] with this property?
 
  • #4
Eulogy said:
Such a sequence could not be totally bounded since if I took a finite cover of the space with balls radius say half then each function in the sequence must be in a separate ball and as the sequence is infinite this leads to a contradiction. But I can't think of a sequence functions that are continuous on [0,1] with this property?

Can you think of a way to define f1 so that it's integral is 1/2, but it is only nonzero on [1/2,1]. Same thing for f2 but it is only nonzero on [1/4,1/2], put f3 on [1/8,1/4] etc.
 
Last edited:
  • #5
Yeah just figured it out thanks for your help!
 

FAQ: Is the subset G totally bounded? - Proving or disproving using relevant theorems

What is a totally bounded subset?

A totally bounded subset is a subset of a metric space that can be covered by a finite number of open balls with a given radius. In other words, the subset is "bounded" in the sense that it does not extend infinitely far in any direction, and "totally" in the sense that it can be completely covered by a finite number of open balls.

How is a totally bounded subset different from a bounded subset?

While a totally bounded subset is a subset that can be covered by a finite number of open balls, a bounded subset is simply a subset that does not extend infinitely far in any direction. This means that a totally bounded subset is always bounded, but a bounded subset may not necessarily be totally bounded.

What is the significance of a totally bounded subset?

A totally bounded subset is significant because it has useful properties in metric spaces. For example, every totally bounded subset is pre-compact, meaning that it has a compact closure. This makes totally bounded subsets useful in analysis and topology.

How can we determine if a subset is totally bounded?

To determine if a subset is totally bounded, we can use the definition and try to cover the subset with a finite number of open balls. Alternatively, we can use the fact that every totally bounded subset is also pre-compact, and check if the closure of the subset is compact.

Can a totally bounded subset be unbounded?

Yes, a totally bounded subset can be unbounded. This may seem counterintuitive, but it is possible for a subset to be totally bounded and still extend infinitely far in all directions. For example, the set of rational numbers is totally bounded but unbounded in the real numbers.

Back
Top