Solving structured linear systems with large displacement rank

  • Authors:
  • Alin Bostan;Claude-Pierre Jeannerod;Éric Schost

  • Affiliations:
  • Algorithms Project, INRIA Paris-Rocquencourt, 78153 Le Chesnay Cedex, France;Arénaire Project, INRIA Rhônes-Alpes, Laboratoire LIP (CNRS, ENSL, INRIA, UCBL), ENS Lyon, France;Computer Science Department, The University of Western Ontario, London, Ontario, Canada

  • Venue:
  • Theoretical Computer Science
  • Year:
  • 2008

Quantified Score

Hi-index 5.23

Visualization

Abstract

Linear systems with structures such as Toeplitz, Vandermonde or Cauchy-likeness can be solved in O@?(@a^2n) operations, where n is the matrix size, @a is its displacement rank, and O@? denotes the omission of logarithmic factors. We show that for such matrices, this cost can be reduced to O@?(@a^@w^-^1n), where @w is a feasible exponent for matrix multiplication over the base field. The best known estimate for @w is @w