Source author record

Giorgio Ottaviani

Explore this author’s publications, collaborators and research topics. This catalog record has not been claimed by its author.

ResearcherUnclaimed source record

math.AG math.AC Computational Complexity math.OC alg-geom math.NA Computer Vision math.CA

Catalog footprint

Research overview

28works

8topics

4close collaborators

Actions

Is this your work?

Open graph Browse papers

preprint2022arXiv

Tensor Rank and Complexity

These lecture notes are intended as an introduction to several notions of tensor rank and their connections to the asymptotic complexity of matrix multiplication. The latter is studied with the exponent of matrix multiplication, which will be expressed in terms of tensor (border) rank, (border) symmetric rank and the asymptotic rank of certain tensors. We introduce the multilinear rank of a tensor as well, deal with the concept of tensor equivalence and study prehomogeneous vector spaces with the castling transform. Moreover, we treat Apolarity Theory and use it to determine the symmetric rank (Waring rank) of some symmetric tensors.

preprint2021arXiv

A footnote to a footnote to a paper of B. Segre

The paper is devoted to a detailed study of sextics in three variables having a decomposition as a sum of nine powers of linear forms. This is the unique case of a Veronese image of the plane which, in the terminology introduced by Ciliberto and the first author in [12], is weakly defective, and non-identifiable. The title originates from a paper of 1981, where Arbarello and Cornalba state and prove a result on plane curves with preassigned singularities, which is relevant to extend the studies of B. Segre on special linear series on curves. We explore the apolar ideal of a sextic $F$ and the associated catalecticant maps, in order to determine the minimal decompositions. A particular attention is played to the postulation of the decompositions. Starting with forms with a decomposition $A$ of length $9$ , the postulation of $A$ determines several loci in the $9$ -secant of the $6$ -Veronese image of $\mathbb P^2$ , which include the lower secant varieties, and the ramification locus, where the decomposition is unique. We prove that equations of all these loci, including the $8$ -th and the $7$ -th secant varieties, are provided by minors of the catalecticant maps and by the invariant $H_{27}$ that we describe in Section 4.

preprint2020arXiv

The Hessian map

In this paper we study the Hessian map $h_{d,r}$ which associates to any hypersurface of degree $d$ in ${\mathbb P}^r$ its Hessian hypersurface. We study general properties of this map and we prove that: $h_{d,1}$ is birational onto its image if $d\geq 5$ ; we study in detail the maps $h_{3,1}$ , $h_{4,1}$ and $h_{3,2}$ ; we study the restriction of the Hessian map to the locus of hypersurfaces of degree $d$ with Waring rank $r+2$ in ${\mathbb P}^r$ , proving that this restriction is injective as soon as $r\geq 2$ and $d\geq 3$ , which implies that $h_{3,3}$ is birational onto its image; we prove that the differential of the Hessian map is of maximal rank on the generic hypersurfaces of degree $d$ with Waring rank $r+2$ in ${\mathbb P}^r$ , as soon as $r\geq 2$ and $d\geq 3$ .

preprint2016arXiv

Effective criteria for specific identifiability of tensors and forms

In applications where the tensor rank decomposition arises, one often relies on its identifiability properties for interpreting the individual rank- $1$ terms appearing in the decomposition. Several criteria for identifiability have been proposed in the literature, however few results exist on how frequently they are satisfied. We propose to call a criterion effective if it is satisfied on a dense, open subset of the smallest semi-algebraic set enclosing the set of rank- $r$ tensors. We analyze the effectiveness of Kruskal's criterion when it is combined with reshaping. It is proved that this criterion is effective for both real and complex tensors in its entire range of applicability, which is usually much smaller than the smallest typical rank. Our proof explains when reshaping-based algorithms for computing tensor rank decompositions may be expected to recover the decomposition. Specializing the analysis to symmetric tensors or forms reveals that the reshaped Kruskal criterion may even be effective up to the smallest typical rank for some third, fourth and sixth order symmetric tensors of small dimension as well as for binary forms of degree at least three. We extended this result to $4 \times 4 \times 4 \times 4$ symmetric tensors by analyzing the Hilbert function, resulting in a criterion for symmetric identifiability that is effective up to symmetric rank $8$ , which is optimal.

preprint2016arXiv

On generic identifiability of symmetric tensors of subgeneric rank

We prove that the general symmetric tensor in $S^d {\mathbb C}^{n+1}$ of rank r is identifiable, provided that r is smaller than the generic rank. That is, its Waring decomposition as a sum of r powers of linear forms is unique. Only three exceptional cases arise, all of which were known in the literature. Our original contribution regards the case of cubics ( $d=3$ ), while for $d\ge 4$ we rely on known results on weak defectivity by Ballico, Ciliberto, Chiantini, and Mella.

preprint2016arXiv

The Chow Form of the Essential Variety in Computer Vision

The Chow form of the essential variety in computer vision is calculated. Our derivation uses secant varieties, Ulrich sheaves and representation theory. Numerical experiments show that our formula can detect noisy point correspondences between two images.

preprint2016arXiv

The Euclidean Distance Degree of Orthogonally Invariant Matrix Varieties

We show that the Euclidean distance degree of a real orthogonally invariant matrix variety equals the Euclidean distance degree of its restriction to diagonal matrices. We illustrate how this result can greatly simplify calculations in concrete circumstances.

preprint2016arXiv

The geometry of rank decompositions of matrix multiplication I: 2x2 matrices

This is the first in a series of papers on rank decompositions of the matrix multiplication tensor. In this paper we: establish general facts about rank decompositions of tensors, describe potential ways to search for new matrix multiplication decompositions, give a geometric proof of the theorem of Burichenko's theorem establishing the symmetry group of Strassen's algorithm, and present two particularly nice subfamilies in the Strassen family of decompositions.

preprint2015arXiv

A Geometric Perspective on the Singular Value Decomposition

This is an introductory survey, from a geometric perspective, on the Singular Value Decomposition (SVD) for real matrices, focusing on the role of the Terracini Lemma. We extend this point of view to tensors, we define the singular space of a tensor as the space spanned by singular vector tuples and we study some of its basic properties.

preprint2015arXiv

On real typical ranks

We study typical ranks with respect to a real variety $X$ . Examples of such are tensor rank ( $X$ is the Segre variety) and symmetric tensor rank ( $X$ is the Veronese variety). We show that any rank between the minimal typical rank and the maximal typical rank is also typical. We investigate typical ranks of $n$ -variate symmetric tensors of order $d$ , or equivalently homogeneous polynomials of degree $d$ in $n$ variables, for small values of $n$ and $d$ . We show that $4$ is the unique typical rank of real ternary cubics, and quaternary cubics have typical ranks $5$ and $6$ only. For ternary quartics we show that $6$ and $7$ are typical ranks and that all typical ranks are between $6$ and $8$ . For ternary quintics we show that the typical ranks are between $7$ and $13$ .

preprint2014arXiv

An algorithm for generic and low-rank specific identifiability of complex tensors

We propose a new sufficient condition for verifying whether generic rank-r complex tensors of arbitrary order admit a unique decomposition as a linear combination of rank-1 tensors. A practical algorithm is proposed for verifying this condition, with which it was established that in all spaces of dimension less than 15000, with a few known exceptions, listed in the paper, generic identifiability holds for ranks up to one less than the generic rank of the space. This is the largest possible rank value for which generic identifiability can hold, except for spaces with a perfect shape. The algorithm can also verify the identifiability of a given specific rank-r decomposition, provided that it can be shown to correspond to a nonsingular point of the r-th order secant variety. For sufficiently small rank, which nevertheless improves upon the known bounds for specific identifiability, some local equations of this variety are known, allowing us to verify this property. As a particular example of our approach, we prove the identifiability of a specific 5x5x5 tensor of rank 7, which cannot be handled by the conditions recently provided in [I. Domanov and L. De Lathauwer, On the Uniqueness of the Canonical Polyadic Decomposition of third-order tensors--Part II: Uniqueness of the overall decomposition, SIAM J. Matrix Anal. Appl. 34(3), 2013]. Finally, we also present a surprising new class of weakly-defective Segre varieties that nevertheless turns out to admit a generically unique decomposition.

preprint2014arXiv

The Euclidean distance degree of an algebraic variety

The nearest point map of a real algebraic variety with respect to Euclidean distance is an algebraic function. For instance, for varieties of low rank matrices, the Eckart-Young Theorem states that this map is given by the singular value decomposition. This article develops a theory of such nearest point maps from the perspective of computational algebraic geometry. The Euclidean distance degree of a variety is the number of critical points of the squared distance to a generic point outside the variety. Focusing on varieties seen in applications, we present numerous tools for exact computations.

preprint2013arXiv

Introduction to the hyperdeterminant and to the rank of multidimensional matrices

This is an introduction to the hyperderminant, according to Gelfand, Kapranov and Zelevinsky. The "triangle inequality", characterizing the Segre varieties such that their dual variety is a hypersurface, is proved in a geometric way (Theorems 3.3 and 6.2). Significant emphasis is given to boundary format case, to pencil of quadrics and to the rank of multidimensional matrices. In the last section we propose three open problems.

preprint2013arXiv

New lower bounds for the border rank of matrix multiplication

The border rank of the matrix multiplication operator for n by n matrices is a standard measure of its complexity. Using techniques from algebraic geometry and representation theory, we show the border rank is at least 2n^2-n. Our bounds are better than the previous lower bound (due to Lickteig in 1985) of 3/2 n^2+ n/2 -1 for all n>2. The bounds are obtained by finding new equations that bilinear maps of small border rank must satisfy, i.e., new equations for secant varieties of triple Segre products, that matrix multiplication fails to satisfy.

preprint2013arXiv

One example of general unidentifiable tensors

The identifiability of parameters in a probabilistic model is a crucial notion in statistical inference. We prove that a general tensor of rank 8 in C^3\otimes C^6\otimes C^6 has at least 6 decompositions as sum of simple tensors, so it is not 8-identifiable. This is the highest known example of balanced tensors of dimension 3, which are not k-identifiable, when k is smaller than the generic rank.

preprint2013arXiv

Refined methods for the identifiability of tensors

We prove that the general tensor of size 2^n and rank k has a unique decomposition as the sum of decomposable tensors if k<= 0.9997 (2^n)/(n+1) (the constant 1 being the optimal value). Similarly, the general tensor of size 3^n and rank k has a unique decomposition as the sum of decomposable tensors if k<= 0.998 (3^n)/(2n+1) (the constant 1 being the optimal value). Some results of this flavor are obtained for tensors of any size, but the explicit bounds obtained are weaker.

preprint2013arXiv

The number of singular vector tuples and uniqueness of best rank one approximation of tensors

In this paper we discuss the notion of singular vector tuples of a complex valued $d$ -mode tensor of dimension m_1 x ... x m_d. We show that a generic tensor has a finite number of singular vector tuples, viewed as points in the corresponding Segre product. We give the formula for the number of singular vector tuples. We show similar results for tensors with partial symmetry. We give analogous results for the homogeneous pencil eigenvalue problem for cubic tensors, i.e. m_1=...=m_d. We show uniqueness of best approximations for almost all real tensors in the following cases: rank one approximation; rank one approximation for partially symmetric tensors (this approximation is also partially symmetric); rank-(r_1,...,r_d) approximation for $d$ -mode tensors.

preprint2012arXiv

A computational approach to Lüroth quartics

A plane quartic curve is called Lüroth if it contains the ten vertices of a complete pentalateral. White and Miller constructed in 1909 a covariant quartic 4-fold, associated to any plane quartic. We review their construction and we show how it gives a computational tool to detect if a plane quartic is Lüroth. As a byproduct, the 28 bitangents of a general plane quartic correspond to 28 singular points of the associated White-Miller quartic 4-fold.

preprint2012arXiv

Matrices with Eigenvectors in a Given Subspace

The Kalman variety of a linear subspace in a vector space consists of all endomorphism that possess an eigenvector in that subspace. We study the defining polynomials and basic geometric invariants of the Kalman variety.

preprint2011arXiv

Equations for secant varieties of Veronese and other varieties

New classes of modules of equations for secant varieties of Veronese varieties are defined using representation theory and geometry. Some old modules of equations (catalecticant minors) are revisited to determine when they are sufficient to give scheme-theoretic defining equations. An algorithm to decompose a general ternary quintic as the sum of seven fifth powers is given as an illustration of our methods. Our new equations and results about them are put into a larger context by introducing vector bundle techniques for finding equations of secant varieties in general. We include a few homogeneous examples of this method.

preprint2011arXiv

Equations for secant varieties to Veronese varieties

We define new classes of modules of equations for secant varieties of Veronese varieties using representation theory and geometry. We also revisit some old modules of equations (catalecticant minors) to determine when they are sufficient to give scheme-theoretic defining equations.

preprint2011arXiv

Laplace Equations and the Weak Lefschetz Property

We prove that r independent homogeneous polynomials of the same degree d become dependent when restricted to any hyperplane if and only if their inverse system parameterizes a variety whose (d-1)-osculating spaces have dimension smaller than expected. This gives an equivalence between an algebraic notion (called Weak Lefschetz Property) and a differential geometric notion, concerning varieties which satisfy certain Laplace equations. In the toric case, some relevant examples are classified and as byproduct we provide counterexamples to Ilardi's conjecture.

preprint2011arXiv

On generic identifiability of 3-tensors of small rank

We introduce an inductive method for the study of the uniqueness of decompositions of tensors, by means of tensors of rank 1. The method is based on the geometric notion of weak defectivity. For three-dimensional tensors of type (a, b, c), a\le b\le c, our method proves that the decomposition is unique (i.e. k-identifiability holds) for general tensors of rank k, as soon as k\le (a+1)(b+1)/16. This improves considerably the known range for identifiability. The method applies also to tensor of higher dimension. For tensors of small size, we give a complete list of situations where identifiability does not hold. Among them, there are 4\times4\times4 tensors of rank 6, an interesting case because of its connection with the study of DNA strings.

preprint2011arXiv

On the Waring problem for polynomial rings

In this note we discuss an analog of the classical Waring problem for C[x_0, x_1,...,x_n]. Namely, we show that a general homogeneous polynomial p \in C[x_0,x_1,...,x_n] of degree divisible by k\ge 2 can be represented as a sum of at most k^n k-th powers of homogeneous polynomials in C[x_0, x_1,...,x_n]. Noticeably, k^n coincides with the number obtained by naive dimension count.

preprint2010arXiv

On singular Luroth quartics

Plane quartics containing the ten vertices of a complete pentalateral and limits of them are called Lüroth quartics. The locus of singular Lüroth quartics has two irreducible components, both of codimension two in $¶^{14}$ . We compute the degree of them and we discuss the consequences of this computation on the explicit form of the Lüroth invariant. One important tool are the Cremona hexahedral equations of the cubic surface. We also compute the class in $\overline{M}_3$ of the closure of the locus of nonsingular Lüroth quartics.

preprint2007arXiv

On partial polynomial interpolation

The Alexander-Hirschowitz theorem says that a general collection of $k$ double points in ${\bf P}^n$ imposes independent conditions on homogeneous polynomials of degree $d$ with a well known list of exceptions. We generalize this theorem to arbitrary zero-dimensional schemes contained in a general union of double points. We work in the polynomial interpolation setting. In this framework our main result says that the affine space of polynomials of degree $\le d$ in $n$ variables, with assigned values of any number of general linear combinations of first partial derivatives, has the expected dimension if $d\neq 2$ with only five exceptional cases. If $d=2$ the exceptional cases are fully described.

preprint1995arXiv

On singularities of ${\cal M}_{I\!\! P^3}(c_1,c_2)$

Let ${\cal M}_{I\!\! P^3}(c_1,c_2)$ be the moduli space of stable rank- $2$ vector bundles on $I\!\! P^3$ with Chern classes $c_1$ , $c_2$ . We prove the following results. 1) Let $0 \le β< γ$ be two integers, ( $γ\ge 2)$ , such that $2γ-3β>0$ ; then ${\cal M}_{I\!\! P^3}(0,2γ^2-3β^2)$ is singular (the case $β=0$ was previously proved by M. Maggesi). 2) Let $0 \le β< γ$ be two odd integers ( $γ\ge 5)$ , such that $2γ-3β+1>0$ ; then ${\cal M}_{I\!\! P^3}(-1,2(γ/2)^2-3(β/2)^2+1/4)$ is singular. In particular ${\cal M}_{I\!\! P^3}(0,5)$ , ${\cal M}_{I\!\! P^3}(-1,6)$ are singular.

preprint1994arXiv

The Tangent Space at a Special Symplectic Instanton Bundle on $P_{2n+1}$

Mathematical instanton bundles on $P_3$ have their analogues in rank-- $2n$ instanton bundles on odd dimensional projective spaces $P_{2n+1}$ . The families of special instanton bundles on these spaces generalize the special 'tHooft bundles on $P_3$ . We prove that for a special symplectic instanton bundle $E$ on $P_{2n+1}$ with $c_2=k$ $h^1End( E) = 4(3n-1) k + (2n-5)(2n-1)$ . Therefore the dimension of the moduli space of instanton bundles grows linearly in $k$ . The main difference with the well known case of $P_3$ is that $h^2End( E)$ is nonzero, in fact we prove that it grows quadratically in $k$ . Special symplectic instanton bundles turn out to be singular points of the moduli space. Such bundles $E$ are $SL(2)$ --invariant and the result is obtained regarding the cohomology groups of $E$ as $SL(2)$ --representations.

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Giorgio Ottaviani

Research overview

Is this your work?

28 publications

Tensor Rank and Complexity

A footnote to a footnote to a paper of B. Segre

The Hessian map

Effective criteria for specific identifiability of tensors and forms

On generic identifiability of symmetric tensors of subgeneric rank

The Chow Form of the Essential Variety in Computer Vision

The Euclidean Distance Degree of Orthogonally Invariant Matrix Varieties

The geometry of rank decompositions of matrix multiplication I: 2x2 matrices

A Geometric Perspective on the Singular Value Decomposition

On real typical ranks

An algorithm for generic and low-rank specific identifiability of complex tensors

The Euclidean distance degree of an algebraic variety

Introduction to the hyperdeterminant and to the rank of multidimensional matrices

New lower bounds for the border rank of matrix multiplication

One example of general unidentifiable tensors

Refined methods for the identifiability of tensors

The number of singular vector tuples and uniqueness of best rank one approximation of tensors

A computational approach to Lüroth quartics

Matrices with Eigenvectors in a Given Subspace

Equations for secant varieties of Veronese and other varieties

Equations for secant varieties to Veronese varieties

Laplace Equations and the Weak Lefschetz Property

On generic identifiability of 3-tensors of small rank

On the Waring problem for polynomial rings

On singular Luroth quartics

On partial polynomial interpolation

On singularities of ${\cal M}_{I\!\! P^3}(c_1,c_2)$

The Tangent Space at a Special Symplectic Instanton Bundle on $P_{2n+1}$

See the researcher in context

Building this map preview

Giorgio Ottaviani

Research overview

Is this your work?

28 publications

Tensor Rank and Complexity

A footnote to a footnote to a paper of B. Segre

The Hessian map

Effective criteria for specific identifiability of tensors and forms

On generic identifiability of symmetric tensors of subgeneric rank

The Chow Form of the Essential Variety in Computer Vision

The Euclidean Distance Degree of Orthogonally Invariant Matrix Varieties

The geometry of rank decompositions of matrix multiplication I: 2x2 matrices

A Geometric Perspective on the Singular Value Decomposition

On real typical ranks

An algorithm for generic and low-rank specific identifiability of complex tensors

The Euclidean distance degree of an algebraic variety

Introduction to the hyperdeterminant and to the rank of multidimensional matrices

New lower bounds for the border rank of matrix multiplication

One example of general unidentifiable tensors

Refined methods for the identifiability of tensors

The number of singular vector tuples and uniqueness of best rank one approximation of tensors

A computational approach to Lüroth quartics

Matrices with Eigenvectors in a Given Subspace

Equations for secant varieties of Veronese and other varieties

Equations for secant varieties to Veronese varieties

Laplace Equations and the Weak Lefschetz Property

On generic identifiability of 3-tensors of small rank

On the Waring problem for polynomial rings

On singular Luroth quartics

On partial polynomial interpolation

On singularities of MI ⁣ ⁣P3(c1,c2){\cal M}_{I\!\! P^3}(c_1,c_2)MIP3​(c1​,c2​)

The Tangent Space at a Special Symplectic Instanton Bundle on P2n+1P_{2n+1}P2n+1​

See the researcher in context

Building this map preview

On singularities of ${\cal M}_{I\!\! P^3}(c_1,c_2)$

The Tangent Space at a Special Symplectic Instanton Bundle on $P_{2n+1}$