On the improvement per iteration in Karmarkar's algorithm for linear programming (Q918862)

From MaRDI portal
Revision as of 09:25, 21 June 2024 by ReferenceBot (talk | contribs) (‎Changed an Item)
scientific article
Language Label Description Also known as
English
On the improvement per iteration in Karmarkar's algorithm for linear programming
scientific article

    Statements

    On the improvement per iteration in Karmarkar's algorithm for linear programming (English)
    0 references
    1990
    0 references
    Let \(\delta_ n(\alpha)\) be the greatest possible guaranteed potential drop at an iteration of Karmarkar's projective method for linear programming problems in dimension n, when the method uses step length parameter \(\alpha\). The paper shows that \(\delta_ n(\alpha)\) equals \(\ln (1+\alpha)-\ln (1-\frac{\alpha}{(n-1)})\) for n sufficiently large.
    0 references
    potential function
    0 references
    Karmarkar's projective method
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references