Linear interval equations: Computing enclosures with bounded relative or absolute overestimation is NP-hard (Q676167): Difference between revisions

From MaRDI portal
RedirectionBot (talk | contribs)
Changed an Item
Import240304020342 (talk | contribs)
Set profile property.
Property / MaRDI profile type
 
Property / MaRDI profile type: MaRDI publication profile / rank
 
Normal rank

Revision as of 00:58, 5 March 2024

scientific article
Language Label Description Also known as
English
Linear interval equations: Computing enclosures with bounded relative or absolute overestimation is NP-hard
scientific article

    Statements

    Linear interval equations: Computing enclosures with bounded relative or absolute overestimation is NP-hard (English)
    0 references
    0 references
    0 references
    1 October 1997
    0 references
    Let \(A\) be a quadratic interval matrix which is strongly regular and which has rational bounds. It is further assumed that for each \(\delta>0\) there exists a polynomial time algorithm which solves the interval equation \(Ax=b\) for any interval vector \(b\) with rational coefficients within relative or absolute accurracy \(\delta\). Then it is shown that \(P=NP\). This result holds also under the additional restriction that \(A\) is symmetric.
    0 references
    0 references
    linear interval equations
    0 references
    NP-hard
    0 references
    inclusion of solution
    0 references
    interval arithmetic
    0 references
    algorithm
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references