On the gap between separating words and separating their reversals (Q1698731): Difference between revisions
From MaRDI portal
Added link to MaRDI item. |
Set profile property. |
||
Property / MaRDI profile type | |||
Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank |
Revision as of 04:21, 5 March 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | On the gap between separating words and separating their reversals |
scientific article |
Statements
On the gap between separating words and separating their reversals (English)
0 references
16 February 2018
0 references
Given two distinct finite words \(x\), \(y\) over the same alphabet and a deterministic finite automaton \(D\), we say that \(D\) \textit{separates} \(x\) and \(y\) if \(D\) accepts the first word but rejects the second. The minimum number of states required for a deterministic finite automaton to separate \(x\) and \(y\) is denoted by \(\operatorname{sep}(x,y)\). In [Lect. Notes Comput. Sci. 6808, 147--157 (2011; Zbl 1341.68087)] \textit{E. D. Demaine} et al. proposed the following open problem: decide whether the difference \[ | \operatorname{sep}(x,y) - \operatorname{sep}(x^R, y^R)| \] is bounded or not, where \(w^R\) stands for the mirror image of a word \(w\). In this paper the author solves the previously open problem showing that the difference is unbounded for alphabets with at least two letters.
0 references
words separation
0 references
finite automata
0 references