Factoring bivariate polynomials using adjoints (Q2437285)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Factoring bivariate polynomials using adjoints
scientific article

    Statements

    Factoring bivariate polynomials using adjoints (English)
    0 references
    0 references
    3 March 2014
    0 references
    The author studies the problem of factoring bivariate polynomials. To describe these results more precisely suppose that \(k\) is a field, that \(\overline{k}\) is an algebraic closure of \(k\), and that \(F \in k[x,y]\) is a bivariate polynomial of degree \(d\). In addition suppose that \(F\) satisfies the condition that the polynomial \(F(0,y)\) is separable of degree \(d\). In this setting the main result of this paper is an algorithm which yields, under suitable additional hypothesis on \(k\), factorizations of \(F\), both over \(k\) and \(\overline{k}\), assuming given the knowledge of a \(k\)-basis of the vector space \(A\) of adjoint polynomials of \(F\). The author then proves a number of results which concern the complexity this algorithm. These results are related both to the complexity of factoring \(F\) given the knowledge of a \(k\)-basis of \(A\), and also the complexity of computing the vector space \(A\) itself. Finally the author considers concrete illustrative examples. To describe some results, contained in this paper, which make the author's algorithm possible let \(\mathcal{C} \subseteq \mathbb{P}^2\) be the plane curve determined by \(F\). Also suppose that \(Z \) is the closed subscheme of \(\mathcal{C}\) defined by the scheme theoretic intersection of the strict transforms, with respect to a weak embedded resolution of \(\mathcal{C}\), of \(\mathcal{C}\) and a suitable line in \(\mathbb{P}^2\). In this notation, a key step in the author's algorithm is to show that the cokernel of the restriction map \[ \alpha : H^0(\mathcal{C},\mathcal{O}_C) \rightarrow H^0(Z, \mathcal{O}_Z) \] can be effectively computed. This, as the author shows, is achieved by the theory of residues. The next key step is to relate the cokernel of \(\alpha\) to the vector space of adjoint polynomials \(A\), that is the vector space of polynomials which determine the adjoint curves of \(\mathcal{C}\).
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    Factorization
    0 references
    Adjoint polynomials
    0 references
    Curves
    0 references
    Singularities
    0 references
    Residues
    0 references
    Cohomology
    0 references
    0 references
    0 references