Binomiality Testing and Computing Sparse Polynomials via Witness Sets

被引:1
作者
Hauenstein, Jonathan D. [1 ]
Matusevich, Laura [2 ]
Peterson, Chris [3 ]
Sherman, Samantha N. [1 ]
机构
[1] Univ Notre Dame, Dept Appl & Computat Math & Stat, Notre Dame, IN 46556 USA
[2] Texas A&M Univ, Dept Math, College Stn, TX 77843 USA
[3] Colorado State Univ, Dept Math, Ft Collins, CO 80523 USA
关键词
Binomial ideals; Sparse polynomials; Numerical algebraic geometry; Witness sets; Macaulay dual spaces; SYSTEMS;
D O I
10.1007/s10013-021-00543-2
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
Sparse polynomials that vanish on algebraic sets are preferred in many computations since they are easy to evaluate and often arise from underlying structure. For example, a monomial vanishes on an algebraic set if and only if the algebraic set is contained in the union of the coordinate hyperplanes. Eisenbud and Sturmfels initiated a detailed study of binomial ideals roughly 25 years ago and showed that they had many special properties including that each component is rational. Given a general point on a component and its tangent space, this paper exploits rationality to develop a local approach that decides if the component is defined by binomials or not. When a component is not defined by binomials, one often is interested in computing sparse polynomials that vanish on the component. Thus, this paper also develops an approach for computing sparse polynomials using a witness set for the component. Our approach relies on using numerical homotopy methods to sample points on the algebraic set along with incorporating multiplicity information using Macaulay dual spaces. If the algebraic set is defined by polynomials with rational coefficients, exactness recovery such as lattice based methods can be used to find exact representations of the sparse polynomials. Several examples are presented demonstrating the new methods.
引用
收藏
页码:653 / 678
页数:26
相关论文
共 30 条
  • [1] Bailey D. H., 1991, SRCTR92066, P1
  • [2] Bates DJ, 2013, SOFTW ENVIRON TOOLS
  • [3] Bates D.J., BERTINI SOFTWARE NUM, DOI [DOI 10.7274/R0H41PB5, /10.7274/R0H41PB5]
  • [4] Recovering Exact Results from Inexact Numerical Data in Algebraic Geometry
    Bates, Daniel J.
    Hauenstein, Jonathan D.
    McCoy, Timothy M.
    Peterson, Chris
    Sommese, Andrew J.
    [J]. EXPERIMENTAL MATHEMATICS, 2013, 22 (01) : 38 - 50
  • [5] Numerical Software to Compute Newton polytopes and Tropical Membership
    Brysiewicz, Taylor
    [J]. MATHEMATICS IN COMPUTER SCIENCE, 2020, 14 (03) : 577 - 589
  • [6] Stable signal recovery from incomplete and inaccurate measurements
    Candes, Emmanuel J.
    Romberg, Justin K.
    Tao, Terence
    [J]. COMMUNICATIONS ON PURE AND APPLIED MATHEMATICS, 2006, 59 (08) : 1207 - 1223
  • [7] Detecting binomiality
    Conradi, Carsten
    Kahle, Thomas
    [J]. ADVANCES IN APPLIED MATHEMATICS, 2015, 71 : 52 - 67
  • [8] Cox D.A., 2011, GRADUATE STUDIES MAT, V124
  • [9] Numerically deciding the arithmetically Cohen-Macaulayness of a projective scheme
    Daleo, Noah S.
    Hauenstein, Jonathan D.
    [J]. JOURNAL OF SYMBOLIC COMPUTATION, 2016, 72 : 128 - 146
  • [10] DAYTON BH, 2005, ISSAC 05, P116