Open main menu
Home
Random
Recent changes
Special pages
Community portal
Preferences
About Wikipedia
Disclaimers
Incubator escapee wiki
Search
User menu
Talk
Dark mode
Contributions
Create account
Log in
Editing
Surreal number
(section)
Warning:
You are not logged in. Your IP address will be publicly visible if you make any edits. If you
log in
or
create an account
, your edits will be attributed to your username, along with other benefits.
Anti-spam check. Do
not
fill this in!
==Gaps and continuity== In contrast to the real numbers, a (proper) subset of the surreal numbers does not have a least upper (or lower) bound unless it has a maximal (minimal) element. Conway defines<ref name=Con01/> a gap as {{math|{{mset| ''L'' {{!}} ''R'' }}}} such that every element of {{math|''L''}} is less than every element of {{math|''R''}}, and <math display=inline> L \cup R = \mathbb{No}</math>; this is not a number because at least one of the sides is a proper class. Though similar, gaps are not quite the same as [[Dedekind cut]]s,{{efn|The definition of a gap omits the conditions of a Dedekind cut that {{math|''L''}} and {{math|''R''}} be non-empty and that {{math|''L''}} not have a largest element, and also the identification of a cut with the smallest element in {{math|''R''}} if one exists.}} but we can still talk about a completion <math display=inline>\mathbb{No}_\mathfrak{D}</math> of the surreal numbers with the natural ordering which is a (proper class-sized) [[linear continuum]].<ref name=RSS15>{{cite arXiv|first1=Simon|last1=Rubinstein-Salzedo|first2=Ashvin|last2=Swaminathan|eprint=1307.7392v3|class=math.CA|date=2015-05-19|title=Analysis on Surreal Numbers}}</ref> For instance there is no least positive infinite surreal, but the gap <math display=block>\{ x : \exists n \in \mathbb N : x < n\mid x : \forall n\in \mathbb N : x > n \}</math> is greater than all real numbers and less than all positive infinite surreals, and is thus the least upper bound of the reals in <math display=inline>\mathbb{No}_\mathfrak{D}</math>. Similarly the gap <math display=inline>\mathbb{On} = \{ \mathbb{No} \mid{} \}</math> is larger than all surreal numbers. (This is an [[Mathematical joke|esoteric pun]]: In the general construction of ordinals, {{mvar|Ξ±}} "is" the set of ordinals smaller than {{mvar|Ξ±}}, and we can use this equivalence to write {{nowrap|{{math|''Ξ±'' {{=}} {{mset| ''Ξ±'' {{!}} }}}}}} in the surreals; <math display=inline>\mathbb{On}</math> denotes the class of ordinal numbers, and because <math display=inline>\mathbb{On}</math> is [[Cofinal (mathematics)|cofinal]] in <math display=inline>\mathbb{No}</math> we have <math display=inline> \{ \mathbb{No} \mid {} \} = \{ \mathbb{On} \mid {} \} = \mathbb{On}</math> by extension.) With a bit of set-theoretic care,{{efn|Importantly, there is no claim that the collection of Cauchy sequences constitutes a class in NBG set theory.}} <math display=inline>\mathbb{No}</math> can be equipped with a topology where the [[open set]]s are unions of open intervals (indexed by proper sets) and continuous functions can be defined.<ref name=RSS15/> An equivalent of [[Cauchy sequence]]s can be defined as well, although they have to be indexed by the class of ordinals; these will always converge, but the limit may be either a number or a gap that can be expressed as <math display=block>\sum_{\alpha\in\mathbb{No}} r_\alpha \omega^{a_\alpha}</math> with {{math|''a''{{sub|''Ξ±''}}}} decreasing and having no lower bound in <math display=inline>\mathbb{No}</math>. (All such gaps can be understood as Cauchy sequences themselves, but there are other types of gap that are not limits, such as {{math|β}} and <math display=inline>\mathbb{On}</math>).<ref name=RSS15/>
Edit summary
(Briefly describe your changes)
By publishing changes, you agree to the
Terms of Use
, and you irrevocably agree to release your contribution under the
CC BY-SA 4.0 License
and the
GFDL
. You agree that a hyperlink or URL is sufficient attribution under the Creative Commons license.
Cancel
Editing help
(opens in new window)