Cross-products LASSO

Luengo, David; Vía, Javier; Monzon, Sandra; Trigano, Tom; Artés‐Rodríguez, Antonio

doi:10.1109/icassp.2013.6638840

Cited by 10 publications

(5 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• Activations are not too close in order to respect the restrictions imposed by the addressed physical problem . Taking into account these constraints, we can express our optimization problem as [13] minimize…”

Section: Problem Solution a Minimization Problem: Cross-products Penalized Cost Functionmentioning

confidence: 99%

“…In [1]- [4] an initial spike train was obtained by means of the LASSO and a post-processing stage was then applied to keep only the strongest activations and remove those that did not respect the negative co-occurrence period. Then, in [13] a novel penalty term, based on the cross-products of the reconstruction coefficients, was added to the LASSO cost function in order to enforce negative co-occurrence. The resulting cross-products LASSO (CP-LASSO) approach resulted in a non-convex optimization problem that was solved using the successive convex approximations (SCA) technique.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Efficient Iteratively Reweighted LASSO Algorithm for Cross-Products Penalized Sparse Solutions

Luengo

Vía

Trigano

2021

2020 28th European Signal Processing Conference (EUSIPCO)

Self Cite

View full text Add to dashboard Cite

In this paper, we describe an efficient iterative algorithm for finding sparse solutions to a linear system. Apart from the well-known L1 norm regularization, we introduce an additional cost term promoting solutions without too-close activations. This additional term, which is expressed as a sum of cross-products of absolute values, makes the problem nonconvex and difficult to solve. However, the application of the successive convex approximations approach allows us to obtain an efficient algorithm consisting in the solution of a sequence of iteratively reweighted LASSO problems. Numerical simulations on randomly generated waveforms and ECG signals show the good performance of the proposed method.Index Terms-sparsity-aware learning, LASSO, sparse coding, non-convex optimization This work has been partly funded by the Spanish government through the KERMES excellence network (ref. TEC2016-81900-REDT).

show abstract

Section: Problem Solution a Minimization Problem: Cross-products Penalized Cost Functionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Efficient Iteratively Reweighted LASSO Algorithm for Cross-Products Penalized Sparse Solutions

Luengo

Vía

Trigano

2021

2020 28th European Signal Processing Conference (EUSIPCO)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Therefore, the burden of computation can be translated from the front‐end signal acquisition device to the back‐end processor. The CS reconstruction algorithms can be mainly divided into three categories: the first is the

ℓ_{0}

‐norm based greedy algorithm [4–7]; the second is the

ℓ_{1}

‐norm based convex optimisation algorithm [8–11]; the third is the non‐convex optimisation algorithm [12–14].…”

Section: Introductionmentioning

confidence: 99%

Optimised two‐dimensional orthogonal matching pursuit algorithm via singular value decomposition

Cheng

Chen

Wang

et al. 2020

IET signal process.

View full text Add to dashboard Cite

The reconstruction algorithm is one of the core issues in applying compressed sensing theory to practical applications. Two‐dimensional orthogonal matching pursuit (2DOMP) algorithm, as an extension of the traditional orthogonal matching pursuit algorithm, can be used directly for the reconstruction of two‐dimensional signals. With 2D separable sampling, the memory requirements and the complexity of 2DOMP are exponentially reduced. However, in 2DOMP algorithm, the requirement of reconstruction matrix is not taken into consideration, merely measurement matrix is used directly. In this study, singular value decomposition is introduced into 2DOMP algorithm, and 2DOMP algorithm based on singular value decomposition (2DOMP‐SVD) is proposed. Singular value decomposition of separable measurement matrices is used to obtain optimised separable reconstruction matrices and optimised measurements. Numerical experiments demonstrate that the proposed 2DOMP‐SVD algorithm can significantly improve the success rate and robustness of reconstruction. Moreover, the separation design of the matrix can satisfy the requirements for both the measurement matrix and the reconstruction matrix individually, and is suitable for general separable linear system.

show abstract

“…This type of approach leads to practical positive results that usually include a number of spurious activations which need a removal procedure before their potential use as physiologically interpretable signals. This is typically performed using a post-processing stage [9,11] or by minimizing a complex non-convex cost function [10].…”

Section: Introductionmentioning

confidence: 99%

Overcomplete Multi-scale Dictionaries for Efficient Representation of ECG Signals

Meltzer

Luengo

Trigano

2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

The electrocardiogram (ECG) was the first biomedical signal subject of extensive digital signal processing techniques. Essentially, the ECG consists of a cyclic sequence of relevant activations embedded into inactivity time sequences combined with interferences and noise. By its nature, it can be subject of representation as a sparse signal. This work describes an efficient method to create overcomplete multi-scale dictionaries that can be used for sparse ECG representation. Whereas most of the proposed methods to date use fixed waveforms that somehow resemble actual ECG shapes, the main innovation in our approach is selecting ECG waveforms recorded from actual patients. A relevant result of our method is the ability to process long lasting recordings from multiple patients. Simulations on patient actual records from Physionet's PTB Diagnostic ECG Database confirm the good performance of the proposed approach.

show abstract

Cross-products LASSO

Cited by 10 publications

References 16 publications

Efficient Iteratively Reweighted LASSO Algorithm for Cross-Products Penalized Sparse Solutions

Efficient Iteratively Reweighted LASSO Algorithm for Cross-Products Penalized Sparse Solutions

Optimised two‐dimensional orthogonal matching pursuit algorithm via singular value decomposition

Overcomplete Multi-scale Dictionaries for Efficient Representation of ECG Signals

Contact Info

Product

Resources

About