Partial elimination ideals and secant cones (Q536032): Difference between revisions

Let \(K\) be an algebraically closed field. Let \(P\subseteq K[x_0, \dots, x_n]\) be a homogeneous maximal ideal and \(S\) the homogeneous subalgebra of \(K[x_0, \dots, x_n]\) generated by the linear forms of \(P\). Let \(I\subset K[x_0, \dots, x_n]\) be a homogeneous ideal and \(y\) a linear form not in \(P\). The \(k\)-th partial elimination ideal of \(I\) with respect to \(P\) is given by \[ K^P_k(I)={\bigoplus}_{d\in\mathbb{Z}}\{f\in S_d\mid\exists g\in (P^{d+1})_{d+k}:y\;^k\!f+g\in I_{d+k}\}. \] It is proved that the cone of \((k+1)\)-secant lines of a closed subscheme \(Z\subseteq\mathbb{P}^n_k\) through a closed point \(p\in\mathbb{P}^n_k\) is defined by the \(k\)-th partial elimination ideal of \(Z\) with respect to \(p\). This is the bases for an algorithm to compute secant cones. Examples are studied using the computer algebra system \textsc{Singular}.