My Math Forum

My Math Forum (http://mymathforum.com/math-forums.php)
-   Number Theory (http://mymathforum.com/number-theory/)
-   -   Limits' and reals' cardinality (http://mymathforum.com/number-theory/343451-limits-reals-cardinality.html)

Loren February 17th, 2018 10:34 PM

Limits' and reals' cardinality
 
Does the limit function relate to a maximum cardinality?

Does the set of real numbers, as they are an "absolute" continuum?

Can one map the set of real numbers onto a finite surface? By bijection?

Micrm@ss February 18th, 2018 07:48 AM

Please start by defining:

1) Limit function. Do you just mean $\lim_{x\rightarrow a} f(x)$?
2) Maximum cardinality
3) Absolute continuum
4) Finite surface

Loren February 19th, 2018 03:54 PM

1) Yes.

2) Does the limit function approach f(a) infinitesimally, e.g. uncountably in decimal form? Might this difference, the limit function minus f(a) itself, be represented by a set of infinite cardinality, since the limit function is actually arbitrary?

3) The set of real numbers, being ["absolute" is redundant and informal, I guess] a continuum, are represented by what cardinality?

4) Can the set of real numbers have a bijection onto any finite surface?

greg1313 February 19th, 2018 07:48 PM

4) How many points are on a finite surface? What is the cardinality of this set?

romsek February 19th, 2018 08:01 PM

This seems to be the sort of thing OP is after.

https://en.wikipedia.org/wiki/Space-filling_curve

Maschke February 19th, 2018 08:25 PM

Quote:

Originally Posted by Loren (Post 588805)
2) Does the limit function approach f(a) infinitesimally, e.g. uncountably in decimal form? Might this difference, the limit function minus f(a) itself, be represented by a set of infinite cardinality, since the limit function is actually arbitrary?

Hi Loren. I'm not sure exactly what you're getting at. Can you provide some context? Limits in the real numbers are defined by quantifying over uncountable sets. When we say, "for every epsilon ..." and then give the rest of the modern formal definition, we are implicitly quantifying over uncountably many potential epsilons. This has philosophical and historical implications when we are talking about how the limit concept replaced the old idea of infinitesimals. But I don't know if that's what your question is getting at, or if you have something else in mind.

Perhaps you can give a specific example of some function you have in mind.

You use the phrase, "this difference, the limit function minus f(a) itself ..." and asked if this is a cardinality. But it's a NUMBER, not a set. [Numbers are ultimately sets, but not in this context!]

For example let $f(x) = 0$ if $x \neq 0$, and $f(0) = 47$. In other words when you input anything other than $0$ into the function, you get back $0$. But when you put in $0$, you get $47$.

Now in this case the limit of $f$ as $x \to 0$ is $0$. But $f(0) = 47$. So the difference of the limit of the function at that point, and the value of the function at that point, is $47$. It's a number.

Does that make sense? Remember, a function doesn't even need to have a value at a given point in order to have a limit there.

Quote:

Originally Posted by Loren (Post 588805)
1)
3) The set of real numbers, being ["absolute" is redundant and informal, I guess] a continuum, are represented by what cardinality?

Well, the cardinality of the continuum is the problem that's been driving set theory since Cantor first asked the question. Our usual axioms don't tell us the cardinality of the real number continuum, and the hunt for new axioms hasn't resolved the question. Some think that asking about the cardinality of the continuum is not a meaningful question. Other set theorists have esoteric approaches such as Ultimate-L and the set-theoretic multiverse. [Those are ideas I've read a little about. I would not want to give the impression I know what they mean].

These are very deep waters. Nobody has any idea what is the cardinality of the continuum. We know it's $2^{\aleph_0}$, but nobody has any idea what cardinal that is.

You asked a great question, that's a fact.


Quote:

Originally Posted by Loren (Post 588805)
4) Can the set of real numbers have a bijection onto any finite surface?

Does this relate to your previous question about space-filling curves? They are generally surjections but not bijections. They hit some points more than once. There's a bijection between the real numbers and the unit square, if that's what you mean by a finite surface. You can get a bijection of the unit interval to the unit square by interleaving the digits of the decimal expansions of the coordinates of the points in the square. Then you can biject the unit interval to the reals. And you have to wave your hands at the .4999... = .5 problem. There are only countably many of those pesky things so you can either ignore them or sit down and figure out how to deal with them explicitly.

Is anything I wrote helpful? I'm not sure what is the intent of your questions.

Joppy February 19th, 2018 08:33 PM

Quote:

Originally Posted by romsek (Post 588807)
This seems to be the sort of thing OP is after.

https://en.wikipedia.org/wiki/Space-filling_curve

Also mentioned here (same OP).

v8archie February 20th, 2018 02:31 AM

Quote:

Originally Posted by Maschke (Post 588808)
Limits in the real numbers are defined by quantifying over uncountable sets.

But limits in other contexts quantify over other sets. For example: the real numbers are limits of sequences of rational numbers. In this context it makes no sense to have any irrational epsilon because they haven't been defined.

The "cardinality of a limit" is simply the cardinality of the set from which you select your epsilon under this definition. But it's not a property that is intrinsic to a limit as far as I'm aware.

zylo February 20th, 2018 09:55 AM

The real numbers are given by the unique limiting sequence (not the limit of the sequence), as n approaches "infinity," of n place decimals, which can be counted for all n.

Cardinality of the reals: countable.

EDIT:
Limits of the reals are a subset of the reals.
.3333...3m, any fixed m, to n places of 3's, has the limit as n approaches infinity, 1/3

v8archie February 20th, 2018 10:28 AM

You are talking the same old unmitigated nonsense again.


All times are GMT -8. The time now is 03:24 PM.

Copyright © 2018 My Math Forum. All rights reserved.