Linear systems on a special rational surface (Q1769982)

From MaRDI portal
Revision as of 09:15, 30 July 2024 by Openalex240730090724 (talk | contribs) (Set OpenAlex properties.)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
scientific article
Language Label Description Also known as
English
Linear systems on a special rational surface
scientific article

    Statements

    Linear systems on a special rational surface (English)
    0 references
    5 April 2005
    0 references
    Given a field \(K\), two families of ideals in \(R=K[x_1,\dots ,x_n]\) are studied in the paper, namely ideals \[ I_\phi = <x_1^{\phi(1)},\dots,x_n^{\phi(1)},(x_1x_2)^{\phi(2)},\dots ,(x_{i_1}\dots x_{i_r})^{\phi(r)},\dots > \] and \[ J_\phi = <x_1^{\phi(1)},\dots ,x_n^{\phi(1)},(x_1+x_2)^{2\phi(2)},\dots ,(x_{i_1}+\dots +x_{i_r})^{r\phi(r)},\dots >, \] both \(2^n-1\) generated, where \(\phi :{\mathbb N}\rightarrow {\mathbb N}\) is either a linear degree function (i.e. \(\phi (r) = l + k(n-r)\), \(k,l > 0\)) or an almost linear degree function (i.e. \(\exists k \in {\mathbb N}\), s.t. \(\phi (r) - \phi (r+1)= k\) or \(k+1\), \(\forall r \in \{1,..,n-1\}\)). Such ideals where studied by \textit{A. Postnikov} and \textit{M. Shapiro} [in: Differential topology, infinite-dimensional Lie algebras and applications, Transl. Ser. 2, Am. Math. Soc. 194, 227--235 (1999; Zbl 0971.53094)] (in order to study the algebra generated by curvature forms on the generalized flag manifold) who conjectured that \(I_\phi\) and \(J_\phi\) have the same Hilbert series when \(\phi\) is a linear degree function, while in the almost linear case they observed that often the same happens, but there are counterexamples. While their work is of combinatorial nature, in this paper, with algebraic geometry means, it is proved that the two Hilbert series are the same in both cases (\(\phi\) linear or almost linear) when \(n=3\) and char\(K\) is either 0 or bigger than \(\sum ^3_{i=1}\phi(i) -2\). Since \(I_\phi\) is a monomial ideal, it is possible to compute a (maybe non minimal) resolution of its, and from it to get its Hilbert series. In order to compute the Hilbert series for \(J_\phi\), the author uses Macaulay's inverse systems approach (classically known as apolarity): the dimension of \(({R\over J_{\phi}})_j\) is equal to the dimension of \(F_j\), where \(F\) is the ideal of a fat point scheme \(Z \subset {\mathbb P}^2\), \(Z=a_1P_1+\dots +a_7P_7\), where \(P_1,P_2,P_3\) are the coordinate points and \(P_4=(1:1:0), P_5=(1:0:1), P_6=(0:1:1), P_7=(1:1:1)\) for opportune values of the \(a_i\)'s. The study of linear systems on the rational surface obtained by blowing up the points allows the author to determine the value of \(\dim F_j\) (which is not the same as if the points were generic) and also to give a geometric description of the deviation of \(\dim F_j\) from the dimension of an ideal generated by generic forms. Moreover, an example is given of an almost linear degree function \(\phi\) for which the Hilbert series of \(I_\phi\) and \(J_\phi\) are the same, but their Betti numbers are not equal.
    0 references
    inverse system
    0 references
    apolarity
    0 references
    0 references

    Identifiers