Hadamard–Hitchcock decompositions: Identifiability and computation
Mathematics of Computation · 11.03.2026 · 95 (362) · 2849-2890
DOI: 10.1090/mcom/4174 ↗Kurz gefasst
Fachartikel in der Zeitschrift Mathematics of Computation, online erschienen am 11.03.2026. Verfasst von Alessandro Oneto und 1 weiteren Autor:innen. Die Arbeit nennt 63 Literaturangaben. Gefördert von: Ministero dell’Istruzione, dell’Università e della Ricerca, Simons Foundation, Onderzoeksraad, KU Leuven.
Abstract (Originalsprache)
A Hadamard–Hitchcock decomposition of a multidimensional array is a decomposition that expresses the latter as a Hadamard product of several tensor rank decompositions. Such decompositions can encode probability distributions that arise from statistical graphical models associated to complete bipartite graphs with one layer of observed random variables and one layer of hidden ones, usually called restricted Boltzmann machines. We establish generic identifiability of Hadamard–Hitchcock decompositions by exploiting the reshaped Kruskal criterion for tensor rank decompositions. An algorithm leveraging existing decomposition algorithms for tensor rank decomposition is introduced for computing a Hadamard–Hitchcock decomposition. Numerical experiments illustrate its computational performance and numerical accuracy in a noiseless setting.
Förderung
- Ministero dell’Istruzione, dell’Università e della Ricerca
- Simons Foundation
- Onderzoeksraad, KU Leuven
Zitieren
APA
Oneto, A., & Vannieuwenhoven, N. (2026). Hadamard–Hitchcock decompositions: Identifiability and computation. Mathematics of Computation, 95(362), 2849-2890. https://doi.org/10.1090/mcom/4174
Vancouver
Oneto A, Vannieuwenhoven N. Hadamard–Hitchcock decompositions: Identifiability and computation. Mathematics of Computation. 2026;95(362):2849-2890. doi:10.1090/mcom/4174
BibTeX
@article{Oneto2026,
author = {Oneto, Alessandro and Vannieuwenhoven, Nick},
title = {Hadamard–Hitchcock decompositions: Identifiability and computation},
journal = {Mathematics of Computation},
year = {2026},
volume = {95},
number = {362},
pages = {2849--2890},
doi = {10.1090/mcom/4174}
}
EUVIMED