In the paper, we address the important problem of tensor decompositions which can be seen as a generalisation of Singular Value Decomposition for matrices. We consider general multilinear and multihomogeneous tensors. We show how to reduce the problem to a truncated moment matrix problem and give a new criterion for flat extension of Quasi-Hankel matrices. We connect this criterion to the commutation characterisation of border bases. A new algorithm is described which applies for general multihomogeneous tensors, extending the approach of J.J. Sylvester on binary forms. An example illustrates the algebraic operations involved in this approach and how the decomposition can be recovered from eigenvector computation.
General tensor decomposition, moment matrices and applications / Bernardi, Alessandra; Jerome, Brachat; Pierre, Comon; Bernard, Mourrain. - In: JOURNAL OF SYMBOLIC COMPUTATION. - ISSN 0747-7171. - 52:(2013), pp. 51-71. [10.1016/j.jsc.2012.05.012]
General tensor decomposition, moment matrices and applications
Bernardi, Alessandra;
2013-01-01
Abstract
In the paper, we address the important problem of tensor decompositions which can be seen as a generalisation of Singular Value Decomposition for matrices. We consider general multilinear and multihomogeneous tensors. We show how to reduce the problem to a truncated moment matrix problem and give a new criterion for flat extension of Quasi-Hankel matrices. We connect this criterion to the commutation characterisation of border bases. A new algorithm is described which applies for general multihomogeneous tensors, extending the approach of J.J. Sylvester on binary forms. An example illustrates the algebraic operations involved in this approach and how the decomposition can be recovered from eigenvector computation.File | Dimensione | Formato | |
---|---|---|---|
paper.pdf
accesso aperto
Tipologia:
Pre-print non referato (Non-refereed preprint)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
482.98 kB
Formato
Adobe PDF
|
482.98 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione