The Hermitian Distance degree of an algebraic variety
Davide Furchì
TL;DR
This work extends Euclidean distance degree theory to the Hermitian setting by introducing the virtual Hermitian Distance degree $\mathrm{vHDdeg}$ and Hermitian Distance degree $\mathrm{HDdeg}$, which count and classify critical points of the Hermitian distance to an algebraic variety. It develops a robust algebraic framework, including the Hermitian critical ideal, Hermitian critical set $H_{\mathbf{u}}$, and the HD/ vHD correspondences, enabling both exact and bound computations for hypersurfaces and parametrized varieties. The paper provides explicit bounds (via Bernstein–Kouchnirenko and mixed volumes), a detailed study of conics, and duality relations with the dual variety $X^{\vee}$, together with a Hermitian distance polynomial $\mathrm{HDpol}_{X,\mathbf{u}}(t^2)$ encoding all distance data. It also connects the Hermitian discriminant with the complex evolute concepts, offering geometric interpretations of where the critical-point count changes. Collectively, these results illuminate the structure of nearest-point problems under Hermitian metrics and extend the toolkit for computational algebraic geometry in complex settings.
Abstract
In this paper we develop an algebraic theory to study the problem of finding the minimum distance point from an algebraic variety with respect to the Hermitian distance function. The theory generalizes the Euclidean Distance degree introduced in arXiv:1309.0049, replacing a positive symmetric bilinear form by a Hermitian form. Various examples are presented to show the robustness of the machineries.
