From basis components to complex structural patterns
- 1 May 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15206149,p. 3228-3232
Abstract
A novel approach is proposed to extract high-rank patterns from multiway data. The method is useful when signals comprise collinear components or complex structural patterns. Alternating least squares and multiplication algorithms are developed for the new model with/without non negativity constraints. Experimental results on synthetic data and real-world dataset confirm the validity of the proposed model and algorithms.Keywords
This publication has 25 references indexed in Scilit:
- Low Complexity Damped Gauss--Newton Algorithms for CANDECOMP/PARAFACSIAM Journal on Matrix Analysis and Applications, 2013
- On Revealing Replicating Structures in Multiway Data: A Novel Tensor Decomposition Approach2012
- Uni-mode and Partial Uniqueness Conditions for CANDECOMP/PARAFAC of Three-Way Arrays with Linearly Dependent LoadingsSIAM Journal on Matrix Analysis and Applications, 2012
- Improved Uniqueness Conditions for Canonical Tensor Decompositions with Linearly Dependent LoadingsSIAM Journal on Matrix Analysis and Applications, 2012
- New ALS Methods With Extrapolating Search Directions and Optimal Step Size for Complex-Valued Tensor DecompositionsIEEE Transactions on Signal Processing, 2011
- Enhanced Line Search: A Novel Method to Accelerate PARAFACSIAM Journal on Matrix Analysis and Applications, 2008
- Projected Gradient Methods for Nonnegative Matrix FactorizationNeural Computation, 2007
- Comparison between ISRA and RLA algorithms. Use of a Wiener Filter based stopping criterionAstronomy and Astrophysics Supplement Series, 1999
- Learning the parts of objects by non-negative matrix factorizationNature, 1999
- An Iterative Image Space Reconstruction Algorthm Suitable for Volume ECTIEEE Transactions on Medical Imaging, 1986