Searcharxiv⌕ Search

arXiv subjects

Alessandra Bernardi

Publications and source records attributed to Alessandra Bernardi.

At least 19 recordsLinked to original sources
arXiv2026

Secant varieties of flag varieties via Schur apolarity

We develop a general first-order theory of Schur apolarity for the study of secant varieties of flag varieties in arbitrary homogeneous embeddings. Extending the classical apolarity--fat-point correspondence for Veronese varieties, we show that in the Schur setting the algebraic square of the apolar ideal need not coincide with the geometric double-point conditions. We introduce a geometric Schur square whose relevant component is the conormal space, yielding a Schur Dual Terracini Lemma. Our construction recovers classical apolarity in the symmetric case. A slot-by-slot Consistency Theorem realizes these intrinsic conditions as multigraded double points. As an application, we determine the dimensions of all secant varieties of $\operatorname{Fl}(1,2;V_n)$ embedded by $\mathcal{O}(1,1)$: the only defective cases are $\sigma_2(\operatorname{Fl}(1,2;V_3))$ and $\sigma_3(\operatorname{Fl}(1,2;V_4))$, both of defect one.

math.AG↗
arXiv2026

Hankel and Multiplication Tensor Completions for Cactus Rank

We show that the Hankel flat extension formulation of the cactus algorithm is equivalent to a completion problem for multiplication tensors of Artinian Gorenstein algebras. The unknown Hankel moments are canonically identified with the undetermined tensor coefficients, and under this identification the symbolic multiplication matrices and their commutation equations coincide. This shows that the usual degree extension formulation is a coordinate realization of a variable extension problem with marked generators. We further use Borel-fixed and squat staircases to reduce the family of candidate basis shapes in the resulting algorithm.

math.AC↗
arXiv2026

Neural Learning of Fast Matrix Multiplication Algorithms: A StrassenNet Approach

Fast matrix multiplication can be described as searching for low-rank decompositions of the matrix--multiplication tensor. We design a neural architecture, \textsc{StrassenNet}, which reproduces the Strassen algorithm for $2\times 2$ multiplication. Across many independent runs the network always converges to a rank-$7$ tensor, thus numerically recovering Strassen's optimal algorithm. We then train the same architecture on $3\times 3$ multiplication with rank $r\in\{19,\dots,23\}$. Our experiments reveal a clear numerical threshold: models with $r=23$ attain significantly lower validation error than those with $r\le 22$, suggesting that $r=23$ could actually be the smallest effective rank of the matrix multiplication tensor $3\times 3$. We also sketch an extension of the method to border-rank decompositions via an $\varepsilon$--parametrisation and report preliminary results consistent with the known bounds for the border rank of the $3\times 3$ matrix--multiplication tensor.

math.AG↗
arXiv2026

Triangular tensor networks, pencils of matrices and beyond

We study tensor network varieties associated with the triangular graph, with a focus on the case where one of the physical dimensions is 2. This allows us to interpret the tensors as pencils of matrices. We provide a complete characterization of these varieties in terms of the Kronecker invariants of pencils. We determine their dimension, identifying the cases for which the dimension is smaller than the expected parameter count. We provide necessary conditions for membership in these varieties, in terms of the geometry of classical determinantal varieties, coincident root loci and plane cubic curves. We address some extensions to arbitrary graphs.

math.AG↗
arXiv2025

A refinement on the local cactus rank algorithm

We present an algorithm to recover a minimal local apolar scheme to a homogeneous polynomial $F$. The socle degree of the scheme determines whether it is evinced by a Generalized Additive Decomposition (GAD) of $F$ or of an extension. We give constructive procedures for both cases and compute the Hilbert function efficiently via Hankel operators.

math.AG↗
arXiv2024

Decomposition loci of tensors

The decomposition locus of a tensor is the set of rank-one tensors appearing in a minimal tensor-rank decomposition of the tensor. For tensors lying on the tangential variety of any Segre variety, but not on the variety itself, we show that the decomposition locus consists of all rank-one tensors except the tangency point only. We also explicitly compute decomposition loci of all tensors belonging to tensor spaces with finitely many orbits with respect to the action of product of general linear groups.

math.AG↗
arXiv2024

A quantum implementation of high-order power method for estimating geometric entanglement of pure states

Entanglement is one of the fundamental properties of a quantum state and is a crucial differentiator between classical and quantum computation. There are many ways to define entanglement and its measure, depending on the problem or application under consideration. Each of these measures may be computed or approximated by multiple methods. However, hardly any of these methods can be run on near-term quantum hardware. This work presents a quantum adaptation of the iterative high-order power method for estimating the geometric measure of entanglement of multi-qubit pure states using rank-1 tensor approximation. This method is executable on early fault-tolerant (hybrid) quantum hardware and does not depend on quantum memory. We simulate this algorithm and mitigate the effects of noise on the results of the computation using a theoretical model based on a known mitigation approach, which assumes a global depolarising noise channel.

quant-ph↗
arXiv2024

On schemes evinced by generalized additive decompositions and their regularity

We define and explicitly construct schemes evinced by generalized additive decompositions (GADs) of a given $d$-homogeneous polynomial $F$. We employ GADs to investigate the regularity of $0$-dimensional schemes apolar to $F$, focusing on those satisfying some minimality conditions. We show that irredundant schemes to $F$ need not be $d$-regular, unless they are evinced by special GADs of $F$. Instead, we prove that tangential decompositions of minimal length are always $d$-regular, as well as irredundant apolar schemes of length at most $2d+1$.

math.AC↗
arXiv2024

On the cactus rank of cubics forms

We prove that the smallest degree of an apolar 0-dimensional scheme of a general cubic form in $n+1$ variables is at most $2n+2$, when $n\geq 8$, and therefore smaller than the rank of the form. For the general reducible cubic form the smallest degree of an apolar subscheme is $n+2$, while the rank is at least $2n$.

math.AG↗
arXiv2023

Tensoring by a plane maintains secant-regularity in degree at least two

Starting from an integral projective variety $Y$ equipped with a very ample, non-special and not-secant defective line bundle $\mathcal{L}$, the paper establishes, under certain conditions, the regularity of $(Y \times \mathbb P^2,\mathcal{L}[t])$ for $t\geq 2$. The mildness of those conditions allow to classify all secant defective cases of any product of $(\mathbb P^1)^{ j}\times (\mathbb P^2)^{k}$, $j,k \geq 0$, embedded in multidegree at least $(2, \ldots , 2)$ and $(\mathbb{P}^m\times\mathbb{P}^n\times (\mathbb{P}^2)^k, \mathcal{O}_{\mathbb{P}^m\times\mathbb{P}^n\times (\mathbb{P}^2)^k} (d,e,t_1, \ldots, t_k))$ where $d,e \geq 3$, $t_i\geq 2$, for any $n$ and $m$.

math.AG↗
arXiv2022

Dimension of Tensor Network varieties

The tensor network variety is a variety of tensors associated to a graph and a set of positive integer weights on its edges, called bond dimensions. We determine an upper bound on the dimension of the tensor network variety. A refined upper bound is given in cases relevant for applications such as varieties of matrix product states and projected entangled pairs states. We provide a range (the "supercritical range") of the parameters where the upper bound is sharp.

quant-ph↗
arXiv2022

A note on the maximal rank

We give an upper-bound for the $X$-rank of points with respect to a non-degenerate irreducible variety $X$ in the case that sub-generic $X$-rank points generate a hypersurface. We give examples where this bound is sharp and it improves the existing ones.

math.AG↗
arXiv2021

Identifiability of rank-3 tensors

Rank-2 and rank-3 tensors are almost all identifiable with only few exceptions. We classify them all together with the dimensions and the structures of all the sets evincing the rank.

math.AG↗
arXiv2020

Waring, tangential and cactus decompositions

(EN) We revise the famous algorithm for symmetric tensor decomposition due to Brachat, Comon, Mourrain and Tsidgaridas. Afterwards, we generalize it in order to detect possibly different decompositions involving points on the tangential variety of a Veronese variety. Finally, we produce an algorithm for cactus rank and decomposition, which also detects the support of the minimal apolar scheme and its length at each component. (FR) Nous revenons sur le fameux algorithme de Brachat, Comon, Mourrain et Tsidgaridas pour la dćomposition des tenseurs symétriques. Ensuite, nous le généralisons afin de détecter de possibles décompositions différentes impliquant des points sur la variété tangentielle d'une variété de Veronese. Enfin, nous proposons un algorithme pour le rang et la décomposition cactus, qui, lui aussi, détecte le support du schéma apolaire minimal ainsi que sa longueur sur chaque composante.

math.AC↗
arXiv2020

Geometric conditions for strict submultiplicativity of rank and border rank

The $X$-rank of a point $p$ in projective space is the minimal number of points of an algebraic variety $X$ whose linear span contains $p$. This notion is naturally submultiplicative under tensor product. We study geometric conditions that guarantee strict submultiplicativity. We prove that in the case of points of rank two, strict submultiplicativity is entirely characterized in terms of the trisecant lines to the variety. Moreover, we focus on the case of curves: we prove that for curves embedded in an even-dimensional projective space, there are always points for which strict submultiplicativity occurs, with the only exception of rational normal curves.

math.AG↗
arXiv2019

High Order Singular Value Decomposition for Plant Biodiversity Estimation

We propose a new method to estimate plant biodiversity with R{é}nyi and Rao indexes through the so called High Order Singular Value Decomposition (HOSVD) of tensors. Starting from NASA multispectral images we evaluate biodiversity and we compare original biodiversity estimates with those realised via the HOSVD compression methods for big data. Our strategy turns out to be extremely powerful in terms of storage memory and precision of the outcome. The obtained results are so promising that we can support the efficiency of our method in the ecological framework.

eess.SP↗
arXiv2019

Strict inclusions of high rank loci

For a given projective variety $X$, the high rank loci are the closures of the sets of points whose $X$-rank is higher than the generic one. We show examples of strict inclusion between two consecutive high rank loci. Our first example is for the Veronese surface of plane quartics. Although Piene had already shown an example when $X$ is a curve, we construct infinitely many curves in $\mathbb P^4$ for which such strict inclusion appears. For space curves, we give two criteria to check whether the locus of points of maximal rank 3 is finite (possibly empty).

math.AG↗