Tensorlab 3.0 — Numerical optimization strategies for large-scale constrained and coupled matrix/tensor factorization

Vervliet, Nico; Debals, Otto; Lathauwer, Lieven De

doi:10.1109/acssc.2016.7869679

Cited by 250 publications

(367 citation statements)

References 20 publications

Supporting

Mentioning

365

Contrasting

Unclassified

Order By: Relevance

“…Recently, dedicated methods have been designed for large, sparse, or incomplete tensors: See for example [7]- [9] and references therein. Also, support for structured and coupled tensor decompositions has been added to tensor toolboxes such as Tensorlab [10], [11].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Nonlinear least squares updating of the canonical polyadic decomposition

Vandecappelle

Vervliet

Lathauwer

2017

2017 25th European Signal Processing Conference (EUSIPCO)

Self Cite

View full text Add to dashboard Cite

Abstract-Current batch tensor methods often struggle to keep up with fast-arriving data. Even storing the full tensors that have to be decomposed can be problematic. To alleviate these limitations, tensor updating methods modify a tensor decomposition using efficient updates instead of recomputing the entire decomposition when new data becomes available. In this paper, the structure of the decomposition is exploited to achieve fast updates for the canonical polyadic decomposition whenever new slices are added to the tensor in a certain mode. A batch NLS-algorithm is adapted so that it can be used in an updating context. By only storing the old decomposition and the new slice of the tensor, the algorithm is both time-and memory efficient. Experimental results show that the proposed method is faster than batch ALS and NLS methods, while maintaining a good accuracy for the decomposition.

show abstract

Section: Introductionmentioning

confidence: 99%

“…We apply the framework for the computation of structured tensor decompositions proposed in [15], [16] to modify the CPD when new tensor slices become available and old slices become outdated. This yields a CPD updating method that is more efficient than the existing batch methods, while maintaining a good accuracy.…”

Section: Introductionmentioning

confidence: 99%

Nonlinear least squares updating of the canonical polyadic decomposition

Vandecappelle

Vervliet

Lathauwer

2017

2017 25th European Signal Processing Conference (EUSIPCO)

Self Cite

View full text Add to dashboard Cite

show abstract

“…There are several tensor toolboxes available for the Matlab platform, like the Matlab Tensor toolbox, see [5] and the Tensorlab, see [38]. In this work, we have used an HOSVD implementation using tensor operations from [5].…”

Section: Algorithms For Tensor Decompositionsmentioning

confidence: 99%

Flow-based dissimilarity measures for reservoir models: a spatial-temporal tensor approach

2017

View full text Add to dashboard Cite

In reservoir engineering, it is attractive to characterize the difference between reservoir models in metrics that relate to the economic performance of the reservoir as well as to the underlying geological structure. In this paper, we develop a dissimilarity measure that is based on reservoir flow patterns under a particular operational strategy. To this end, a spatial-temporal tensor representation of the reservoir flow patterns is used, while retaining the spatial structure of the flow variables. This allows reducedorder tensor representations of the dominating patterns and simple computation of a flow-induced dissimilarity measure between models. The developed tensor techniques are applied to cluster model realizations in an ensemble, based on similarity of flow characteristics.

show abstract

“…To estimate the number of components required to perform the decomposition, R values were estimated with the MatLab toolbox Tensorlab [11]. First, using the rankest function of Tensorlab, an upper bound of R = 4 and R = 9 was obtained for HH tensors and W tensors, respectively.…”

Section: Tensor Decompositionmentioning

confidence: 99%

Nonconvulsive epileptic seizures detection using multiway data analysis

Aldana

Hunyadi

Reyes

et al. 2017

2017 25th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

Abstract-Nonconvulsive status epilepticus (NCSE) is observed when the patient undergoes a persistent electroencephalographic epileptic episode without physical symptoms. This condition is commonly found in critically ill patients from intensive care units and constitutes a medical emergency. This paper proposes a method to detect nonconvulsive epileptic seizures (NCES). To perform the NCES detection the electroencephalogram (EEG) is represented as a third order tensor with axes f requency × time × channels using Wavelet or Hilbert-Huang transform. The signatures obtained from the tensor decomposition are used to train five classifiers to separate between the normal and seizure EEG. Classification is performed in two ways: (1) with each signature of the different modes separately, (2) with all signatures assembled. The algorithm is tested on a database containing 139 nonconvulsive seizures. From all performed analysis, Hilbert-Huang Tensors Space and assembled signatures demonstrate to be the best features to classify between seizure and non-seizure EEG.

show abstract

Tensorlab 3.0 — Numerical optimization strategies for large-scale constrained and coupled matrix/tensor factorization

Cited by 250 publications

References 20 publications

Nonlinear least squares updating of the canonical polyadic decomposition

Nonlinear least squares updating of the canonical polyadic decomposition

Flow-based dissimilarity measures for reservoir models: a spatial-temporal tensor approach

Nonconvulsive epileptic seizures detection using multiway data analysis

Contact Info

Product

Resources

About