A counterexample to a conjecture on simultaneous Waring identifiability

DOI10.1016/J.JSC.2023.102223MaRDI QIDQ6170810zbMATH OpenOpenAlexWikidataFDO

Authors Elena Angelini

Publication date 10 August 2023

Published in Journal of Symbolic Computation (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2304.03186

zbMATH Keywords

normal formssimultaneous decompositions

Mathematics Subject Classification ID

Secant varieties, tensor rank, varieties of sums of powers (14N07)

Abstract: The new identifiable case appeared in cite{AGMO}, together with the analysis on simultaneous identifiability of pairs of ternary forms recently developed in cite{BG}, suggested the following conjecture towards a complete classification of all simultaneous Waring identifiable cases: for any $dgeq2$, the general polynomial vectors consisting of $d-1$ ternary forms of degree $d$ and a ternary form of degree $d+1$, with rank $frac{d}^2+d+22$, are identifiable over $mathbfC$. In this paper, by means of a computer-aided procedure inspired to the one described in cite{AGMO}, we obtain that the case $d=4$ contradicts the previous conjecture, admitting at least $36$ complex simultaneous Waring decompositions (of length $11$) instead of $1$.

Review text

A simultaneous Waring decomposition of a set of homogeneous polynomials \(S=\{f_1,\dots,f_k\}\), with \(\deg(f_i)=d_i\), is a set of (homogeneous) linear forms \(L_1, \dots,L_r\) such that each \(f_i\) can be written as a linear combination of \(d_i\)-th powers of the \(L_j\)'s. The author considers the problem of the uniqueness of a minimal simultaneous Waring decomposition (in other words, the identifiability) for a general set \(S\) of polynomials of fixed degrees. Some recent results on the topic suggest the existence of cases in which generic identifiability holds. The author disproves one of these cases, i.e. the case where \(f_1,\dots,f_{k-1}\) are ternary polynomiuals of degree \(k\) and \(f_k\) is a ternary polynomial of degree \(k+1\). A geometric reduction links the identifiability property of \(S\) to properties of secant varieties of a suitable projective bundle \(X\subset \mathbb P^N\). The author constructs an algorithm, implemented in Macaulay2 equipped with the package Bertini. hence based on homotopic continuation, that computes number of secant spaces to \(X\) passing through a generic point of \(\mathbb P^N\). Running the program, the author proves that a set of three generic ternary polynomials of degree \(4\) and one generic ternary polynomial of degree \(5\) has at least \(36\) (minimal) decompositions in terms of \(11\) linear forms.

Recommendations

• Generic identifiability of pairs of ternary forms
• On the number of Waring decompositions for a generic polynomial vector
• Waring decompositions and identifiability via Bertini and Macaulay2 software
• On the identifiability of ternary forms
• On the description of identifiable quartics

Cites work

• Generic identifiability of pairs of ternary forms
• Homotopy techniques for tensor decomposition and perfect identifiability
• scientific article; zbMATH DE number 5968745 (Why is no real title available?)
• scientific article; zbMATH DE number 773851 (Why is no real title available?)
• Identifiability of homogeneous polynomials and Cremona transformations
• Numerically solving polynomial systems with Bertini
• On the number of Waring decompositions for a generic polynomial vector
• Varieties with minimal secant degree and linear systems of maximal dimension on surfaces
• Waring decompositions and identifiability via Bertini and Macaulay2 software
• Weakly defective varieties

Cited in

(1)

• Waring identifiability for powers of forms via degenerations

This page was built for publication: A counterexample to a conjecture on simultaneous Waring identifiability

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6170810)