An intersection problem for finite automata (Q1118410): Difference between revisions
From MaRDI portal
Set profile property. |
ReferenceBot (talk | contribs) Changed an Item |
||
Property / cites work | |||
Property / cites work: Q3922646 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Completeness, invariance and <i>λ</i>-definability / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Logical relations and the typed λ-calculus / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q3795665 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q3760508 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: The lambda calculus, its syntax and semantics / rank | |||
Normal rank |
Revision as of 13:49, 19 June 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | An intersection problem for finite automata |
scientific article |
Statements
An intersection problem for finite automata (English)
0 references
1988
0 references
Let M(A,S) be the set of all initial finite automata with state space S of k states and alphabet A of n letters and f: M(A,S)\(\to S\). The language accepted by \(M\in M(A,S)\) and f is the set of words in A mapping the initial state of M to f(M). A map f satisfies the m-intersection condition if for any m machines in M(A,S) there is a word accepted by all of them. Let m(n,k) be the minimum m such that for any f satisfying the m-intersection condition there is a word which belongs to all the languages accepted by M(A,S) and f. The authors show that m(n,k)\(\geq 3\) for n,k\(\geq 2\) and \(m(2,k)=3\), and give some other partial results. The problem is close to a question by Plotkin concerning the connection between combinators and logical relations.
0 references
initial finite automata
0 references
intersection condition
0 references
combinators
0 references
logical relations
0 references