2

I wanted to know the approximate sum of real numbers from 0 to 1.

Please tell me how we can find it.

Zev Chonoles
  • 129,973
satyatech
  • 517
  • 3
    there are infinitely many(uncountable) numbers greater than $1/2.$ – abel Jun 28 '15 at 17:05
  • 2
    It is difficult to take a sum of an uncountable number of numbers. Even if you restricted the sum to positive rational numbers in top half of the interval, you would get an infinite result. – Henry Jun 28 '15 at 17:05
  • 1
    Something like $\int_{x=0}^1 x$ is more appropriate, I think. – davcha Jun 28 '15 at 17:13

2 Answers2

15

The sum diverges: $$\large\sum_{a\in[0,1]}a=\sup_{\substack{\text{finite}\\S\subset[0,1]}}\sum_{a\in S}a\geq \sup_{N\in\mathbb{N}}\sum_{n=2}^N\left(\frac{1}{2}+\frac{1}{n}\right)\geq\sup_{N\in\mathbb{N}}\left(\frac{N}{2}\right)=\infty$$

Observe this definition from Folland's real analysis book:

enter image description here

Zev Chonoles
  • 129,973
  • 1
    Under this definition with $f(x) \geq 0$ for all $x \in X$, we get $\sum_{x \in X} f(x) = \infty$ whenever there are uncountably many $x \in X$ such that $f(x)>0$. – Michael Jun 28 '15 at 17:17
  • 1
    Indeed, that's Folland's next statement :) (image link) – Zev Chonoles Jun 28 '15 at 17:20
  • Interesting. I have never seen such a definition. Perhaps it is only useful for that fact itself...which can be interpeted as saying that the sum is always infinity except at times when standard definitions of sums apply. =) Yet your quote says "We shall occasionally encounter uncountable sums," I wonder when such things would actually be encountered. – Michael Jun 28 '15 at 20:27
  • @Michael Re: I wonder when such things would actually be encountered. I think that this might be a good question on the main site. (However, there were already some posts about this type of sum, so if you decide to post a separate question about this, you should search first.) I do not know where they are used in Folland's book. – Martin Sleziak Jul 02 '15 at 08:00
  • 1
    One application I have seen is the definition of the Hilbert space $\ell_2(A)$. If you already learned about Hilbert spaces, you have certainly seen $\ell_2=\ell_2(\mathbb N)$, where the inner product is given by $\langle x,y \rangle = \sum\limits_{i\in\mathbb N} x_iy_i$. If we allow summation over arbitrary sets, then we can define $\ell_2(A)$ using almost the same construction; in this case, the inner product will be $\langle x,y \rangle = \sum\limits_{i\in A} x_iy_i$. It can be shown that this is indeed a Hilbert space and that every space is isomorphic to $\ell_2(A)$ for some set $A$. – Martin Sleziak Jul 02 '15 at 08:01
  • So this is, in some sense, a classification of all Hilbert spaces. This result can be found, for example, in Chapter 13 Roman's Advanced Linear Algebra" or in Chapter 9 of Dixmier's General Topology. You should probably be able to find some other places where this result is shown. – Martin Sleziak Jul 02 '15 at 08:01
  • @Michael Only after posting the above comments I have found this question which asks exactly about applications of uncountable sums: Does uncountable summation, with a finite sum, ever occur in mathematics? – Martin Sleziak Jul 02 '15 at 08:05
6

The sum is infinite. We don't really have a good definition for $$\sum_{x\in [0,1]}x,$$ But it seems reasonable that he following relation should hold for any proposed definition. $$\sum_{x\in\mathbb [0,1]} x\geq \sum_{n\in \mathbb N} \frac{1}{n}=\infty.$$

Plutoro
  • 22,730
  • 3
  • 41
  • 71