The latent structure of four-dimensional tensors can be investigated by means of the four-way CANDECOMP/PARAFAC model. This technique is seldom used because its estimating design is challenging from an algorithmic and interpretational stand- point. Parameter estimation with a least-squares approach can be computationally costly, especially under difficult conditions such as factor collinearity and model over-specification. In this work, we implement a 4th-order extension of the efficient trilinear procedure INT-2 to tackle estimating setbacks and test it in a simulation study.

On four-way CP model estimation efficiency

Simonacci, Violetta
;
Gallo, Michele
2022-01-01

Abstract

The latent structure of four-dimensional tensors can be investigated by means of the four-way CANDECOMP/PARAFAC model. This technique is seldom used because its estimating design is challenging from an algorithmic and interpretational stand- point. Parameter estimation with a least-squares approach can be computationally costly, especially under difficult conditions such as factor collinearity and model over-specification. In this work, we implement a 4th-order extension of the efficient trilinear procedure INT-2 to tackle estimating setbacks and test it in a simulation study.
2022
Inglese
20
https://link.springer.com/article/10.1007/s00180-022-01271-y
Esperti anonimi
AQLD · CANDECOMP/PARAFAC · Computational efficiency · Multi- way data · QALS · 4th-order tensor
Internazionale
no
2
Simonacci, Violetta; Gallo, Michele
info:eu-repo/semantics/article
262
1 Contributo su Rivista::1.1 Articolo in rivista
open
File in questo prodotto:
File Dimensione Formato  
s00180-022-01271-y.pdf

accesso aperto

Tipologia: Documento in Post-print
Licenza: PUBBLICO - Pubblico con Copyright
Dimensione 1.32 MB
Formato Adobe PDF
1.32 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11574/209858
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
social impact