On the Stable Euclidean Distance Degree of Algebraic Layers

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Graziani, Giacomo
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915747801858048
author Graziani, Giacomo
author_facet Graziani, Giacomo
contents We study the projective geometry of algebraic neural layers, namely families of maps induced by a polynomial activation function, with particular emphasis on the generic Euclidean Distance degree ($\mathrm{gED}$). This invariant is projective in nature and measures the number of optimal approximations of a general point in the ambient space with respect to a general metric. For a fixed architecture (i.e. fixed width and activation polynomial), we prove that the $\mathrm{gED}$ is stably polynomial in the dimensions of the input and output spaces. Moreover, we show that this stable polynomial depends only on the degree of the activation function. Our approach relies on standard intersection theory on the Nash blow-up, which allows us to express the $\gED$ as an intersection number over products of Grassmannians. Stable polynomiality is deduced via equivariant localization, while the reduction to the monomial case follows from an explicit Schubert calculus computation on Grassmannians.
format Preprint
id arxiv_https___arxiv_org_abs_2601_16071
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle On the Stable Euclidean Distance Degree of Algebraic Layers
Graziani, Giacomo
Algebraic Geometry
14N10, 14M15, 68T07, 14C17
We study the projective geometry of algebraic neural layers, namely families of maps induced by a polynomial activation function, with particular emphasis on the generic Euclidean Distance degree ($\mathrm{gED}$). This invariant is projective in nature and measures the number of optimal approximations of a general point in the ambient space with respect to a general metric. For a fixed architecture (i.e. fixed width and activation polynomial), we prove that the $\mathrm{gED}$ is stably polynomial in the dimensions of the input and output spaces. Moreover, we show that this stable polynomial depends only on the degree of the activation function. Our approach relies on standard intersection theory on the Nash blow-up, which allows us to express the $\gED$ as an intersection number over products of Grassmannians. Stable polynomiality is deduced via equivariant localization, while the reduction to the monomial case follows from an explicit Schubert calculus computation on Grassmannians.
title On the Stable Euclidean Distance Degree of Algebraic Layers
topic Algebraic Geometry
14N10, 14M15, 68T07, 14C17
url https://arxiv.org/abs/2601.16071