On the Hilbert function of general fat points in \(\mathbb{P}^1\times\mathbb{P}^1\) (Q2225030)

In the paper under review the authors study the so-called interpolation problem for fat point schemes in \(\mathbb{P}^{1} \times \mathbb{P}^{1}\). A very natural motivation for this study is the celebrated Alexander-Hirschowitz theorem which provides us a complete classification of ideals of double points in general position in the \(n\)-dimensional complex projective space. In particular, this classification contains all the cases when double points fail to impose independent conditions on hypersurfaces of some degree. Let \(S = \mathbb{C}[x_{0},x_{1},y_{0},y_{1}] = \bigoplus_{i,j} S_{i,j}\) be the bigraded coordinate ring of \(\mathbb{P}^{1} \times \mathbb{P}^{1}\). Let \(\mathcal{P} = \{P_{1}, \dots, P_{s}\}\) be a set of points in \(\mathbb{P}^{1} \times \mathbb{P}^{1}\) in general position and for each point \(P_{i}\) we denote by \(\mathfrak{p}_{i}\) the prime ideal defining that point. The scheme of fat points of multiplicity \(m\geq 1\) with the support at \(\mathcal{P}\) is the scheme \(\mathbb{X}\) defined by the ideal \(I_{\mathbb{X}} = \mathfrak{p}_{1}^{m} \cap \dots \cap \mathfrak{p}_{s}^{m}\). Now for any bihomogeneous ideal \(I\) in \(S\), we define the Hilbert function of \(S/I\) as \[\mathrm{HF}_{S/I}(a,b) : = \dim_{\mathbb{C}}(S/I)_{(a,b)} = \dim_{\mathbb{C}}S_{(a,b)} -\dim_{\mathbb{C}} I_{(a,b)} \quad \text{for} \quad (a,b) \in \mathbb{N}^{2}.\] For short, we denote by \(\mathrm{HF}_{\mathbb{X}}\) be the Hilbert function of the quotient ring \(S / I_{\mathbb{X}}\). The key question which gives a motivation for the paper under review can be formulated as follows. Question: Let \(\mathbb{X}\) be a scheme of fat points of multiplicity \(m\) in \(\mathbb{P}^{1} \times \mathbb{P}^{1}\). What is the bigraded Hilbert function of \(\mathbb{X}\)? The main contribution of the present paper is the following general result. Theorem A. Let \(a\geq b\) and we assume that \(b\geq m\). Consider the fat point scheme \(\mathbb{X} = mP_{1} +\dots+ mP_{s} \subset \mathbb{P}^{1} \times \mathbb{P}^{1}\). Then \[\mathrm{HF}_{\mathbb{X}}(a,b) =\min \bigg\{ (a+1)(b+1), s\binom{m+1}{2} -s\binom{m-b}{2}\bigg\},\] except if \(s = 2k+1\) and \(a=bk+c+s(m-b)\) with \(c = 0,\dots, b-2\), where \[\mathrm{HF}_{\mathbb{X}}(a,b) = (a+1)(b+1) - \binom{c+2}{2}.\] Moreover, in the case of triple points, the authors are able to provide a complete description of the associated Hilbert functions. Theorem B. Let \(\mathbb{X} = 3P_{1} + \dots + 3P_{s} \subset \mathbb{P}^{1} \times \mathbb{P}^{1}\). Then \[\mathrm{HF}_{\mathbb{X}}(a,b) =\min\{(a+1)(b+1),6s\},\] except for the following situations: 1) \(b=1\) and \(s < \frac{2}{5}(a+1)\), where \(\mathrm{HF}_{\mathbb{X}}(a,1) = 5s\); 2) \(s=2k+1\) and i) \((a,b) =(4k+1,2)\), where \(\mathrm{HF}_{\mathbb{X}}(4k+1,2) = (a+1)(b+1) - 1\); ii) \((a,b) = (3k,3)\), where \(\mathrm{HF}_{\mathbb{X}}(3k,3) = (a+1)(b+1)-1\); iii) \((a,b) = (3k+1,3)\), where \(\mathrm{HF}_{\mathbb{X}}(3k+1,3) = 6s-1\); 3) \(s=5\) and \((a,b)=(5,4)\), where \(\mathrm{HF}_{\mathbb{X}}(5,4)=29\).