CANDECOMP/PARAFAC Decomposition of High-Order Tensors Through Tensor Reshaping
- 18 June 2013
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 61 (19) , 4847-4860
Abstract
In general, algorithms for order-3 CANDECOMP/ PARAFAC (CP), also coined canonical polyadic decomposition (CPD), are easy to implement and can be extended to higher order CPD. Unfortunately, the algorithms become computationally demanding, and they are often not applicable to higher order and relatively large scale tensors. In this paper, by exploiting the uniqueness of CPD and the relation of a tensor in Kruskal form and its unfolded tensor, we propose a fast approach to deal with this problem. Instead of directly factorizing the high order data tensor, the method decomposes an unfolded tensor with lower order, e.g., order-3 tensor. On the basis of the order-3 estimated tensor, a structured Kruskal tensor, of the same dimension as the data tensor, is then generated, and decomposed to find the final solution using fast algorithms for the structured CPD. In addition, strategies to unfold tensors are suggested and practically verified in the paper.Keywords
All Related Versions
This publication has 54 references indexed in Scilit:
- Generating Correlation Matrices Based on the Boundaries of Their CoefficientsPLOS ONE, 2012
- Tensor decompositions for feature extraction and classification of high dimensional datasetsNonlinear Theory and Its Applications, IEICE, 2010
- Recent advances on the use of separated representationsInternational Journal for Numerical Methods in Engineering, 2009
- An efficient algorithm for Parafac with uncorrelated mode‐A components applied to large I × J × K data sets with I >> JKJournal of Chemometrics, 2009
- Using the simultaneous generalized Schur decomposition as a Candecomp/Parafac algorithm for ill‐conditioned dataJournal of Chemometrics, 2009
- On the Non-Existence of Optimal Solutions and the Occurrence of “Degeneracy” in the CANDECOMP/PARAFAC ModelPsychometrika, 2008
- Cramer-Rao lower bounds for low-rank decomposition of multidimensional arraysIEEE Transactions on Signal Processing, 2001
- On the uniqueness of multilinear decomposition ofN-way arraysJournal of Chemometrics, 2000
- Construction and analysis of degenerate PARAFAC modelsJournal of Chemometrics, 2000
- Error analysis of the generalized rank annihilation methodJournal of Chemometrics, 1994