Talk:Counting problem (complexity)
| This article is rated Stub-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | |||||||||||||||
| |||||||||||||||
Why is y less than or equal to c_r?
[edit ]y, to my understanding, isn't necessarily a number, so what's the meaning of the inequality? I think it's supposed to be y is a subset of the set defined in the cardinality of c_r..
In my understanding the symbol {\displaystyle y} denotes a non-negative integer that can hence be compared to {\displaystyle c_{R}(x)}, the size (that is number of elements) of the set of solutions {\displaystyle \{y\mid (x,y)\in R\}}. I believe one needs to be a bit more careful in phrasing the definition, because a priori there is no bound on the number of solutions since in general {\displaystyle R} is a binary relation between infinite sets (e.g. all strings over a finite (usually two-element) alphabet). Of course one may allow infinitely many solution for certain inputs {\displaystyle x}. For such {\displaystyle x} every pair {\displaystyle (x,n)} will belong to {\displaystyle \#R}. One way to make the number {\displaystyle c_{R}(x)} always finite is to just postulate this in the definition of {\displaystyle R}. Often this is achieved by making more restrictive assumptions such a that there is a function {\displaystyle f\colon \mathbb {N} \to \mathbb {N} } (of a certain kind, e.g. a polynomial function) such that for every input {\displaystyle x} there are only solutions {\displaystyle y} satisfying {\displaystyle (x,y)\in R} with bounded length {\displaystyle |y|\leq f(|x|)} and hence finitely many.
Notation
[edit ]I think the problem needs to be stated in English, as with this unspecified notation it's basically unintelligible. 81.2.154.16 (talk) 20:30, 7 July 2022 (UTC) [reply ]