EXTRAPOLATED ALTERNATING ALGORITHMS FOR APPROXIMATE CANONICAL POLYADIC DECOMPOSITION - Archive ouverte HAL Access content directly
Conference Papers Year : 2020

EXTRAPOLATED ALTERNATING ALGORITHMS FOR APPROXIMATE CANONICAL POLYADIC DECOMPOSITION

Andersen Man Shun Ang
  • Function : Author
Andersen Man
  • Function : Author
Shun Ang
  • Function : Author
Le Thi Khanh
  • Function : Author

Abstract

Tensor decompositions have become a central tool in machine learning to extract interpretable patterns from multiway arrays of data. However, computing the approximate Canonical Polyadic Decomposition (aCPD), one of the most important tensor decomposition model, remains a challenge. In this work, we propose several algorithms based on extrapolation that improve over existing alternating methods for aCPD. We show on several simulated and real data sets that carefully designed extrapolation can significantly improve the convergence speed hence reduce the computational time, especially in difficult scenarios.
Fichier principal
Vignette du fichier
ACHG_ICASSP.pdf (382.99 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-02330641 , version 1 (24-10-2019)
hal-02330641 , version 2 (24-09-2020)

Identifiers

Cite

Andersen Man Shun Ang, Andersen Man, Shun Ang, Jérémy E Cohen, Le Thi Khanh, et al.. EXTRAPOLATED ALTERNATING ALGORITHMS FOR APPROXIMATE CANONICAL POLYADIC DECOMPOSITION. ICASSP, May 2020, Barcelone, Spain. ⟨10.1109/ICASSP40776.2020.9053849⟩. ⟨hal-02330641v2⟩
74 View
306 Download

Altmetric

Share

Gmail Facebook X LinkedIn More