Sorry, you need to enable JavaScript to visit this website.

ICASSP 2020 presentation slide of 'EXTRAPOLATED ALTERNATING ALGORITHMS FOR APPROXIMATE CANONICAL POLYADIC DECOMPOSITION'

Primary tabs

Citation Author(s):
Andersen M. S. Ang, Jérémy Emile Cohen, Le Thi Khanh Hien, Nicolas Gillis
Submitted by:
Man Shun Ang
Last updated:
13 May 2020 - 5:42pm
Document Type:
Presentation Slides
Document Year:
2020
Event:
Presenters Name:
EXTRAPOLATED ALTERNATING ALGORITHMS FOR APPROXIMATE CANONICAL POLYADIC DECOMPOSITION
Paper Code:
WE3.L6.1

Abstract 

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.

up
1 user has voted: Man Shun Ang

Dataset Files

Slide of ICASSP2020 presentation

(133)