Disc-GLasso: Discriminative graph learning with sparsity regularization

Kao, Jiun-Yu; Tian, Dong; Mansour, Hassan; Ortega, Antonio; Vetro, Anthony

doi:10.1109/icassp.2017.7952698

Cited by 13 publications

(44 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The improvement is due to the exploitation of graph structure as side information. For future work, we apply recent graph learning techniques [33,34] to improve the estimation of the underlying connectivity graph. Moreover, we address the problem of deploying the networks for real-time analysis in brain computer interface applications.…”

Section: Resultsmentioning

confidence: 99%

Deep neural networks on graph signals for brain imaging analysis

Guo

Nejati

Cheung

2017

2017 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Brain imaging data such as EEG or MEG are high-dimensional spatiotemporal data often degraded by complex, non-Gaussian noise. For reliable analysis of brain imaging data, it is important to extract discriminative, low-dimensional intrinsic representation of the recorded data. This work proposes a new method to learn the lowdimensional representations from the noise-degraded measurements. In particular, our work proposes a new deep neural network design that integrates graph information such as brain connectivity with fully-connected layers. Our work leverages efficient graph filter design using Chebyshev polynomial and recent work on convolutional nets on graph-structured data. Our approach exploits graph structure as the prior side information, localized graph filter for feature extraction and neural networks for high capacity learning. Experiments on real MEG datasets show that our approach can extract more discriminative representations, leading to improved accuracy in a supervised classification task.

show abstract

Section: Resultsmentioning

confidence: 99%

Deep neural networks on graph signals for brain imaging analysis

Guo

Nejati

Cheung

2017

2017 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

show abstract

“…An interesting open problem is whether the proposed approach can solve generalizations of the 1 -norm minimization problem, such as the graphical lasso (GLASSO) problem [11,34,37].…”

Section: Discussionmentioning

confidence: 99%

A Laplacian approach to $$\ell _1$$-norm minimization

Bonifaci

2021

Comput Optim Appl

View full text Add to dashboard Cite

We propose a novel differentiable reformulation of the linearly-constrained $$\ell _1$$ ℓ 1 minimization problem, also known as the basis pursuit problem. The reformulation is inspired by the Laplacian paradigm of network theory and leads to a new family of gradient-based methods for the solution of $$\ell _1$$ ℓ 1 minimization problems. We analyze the iteration complexity of a natural solution approach to the reformulation, based on a multiplicative weights update scheme, as well as the iteration complexity of an accelerated gradient scheme. The results can be seen as bounds on the complexity of iteratively reweighted least squares (IRLS) type methods of basis pursuit.

show abstract

“…Bandlimited graph signals admit a sparse representation in the graph spectral domain. By considering graph signals as random vectors drawn from a Gaussian Markov random field distribution, the graph learning problem becomes the estimation of the inverse covariance matrix [7,8].…”

Section: Introductionmentioning

confidence: 99%

“…While studies such as [3,4] focus on learning the underlying graph that is efficient with respect to the signal representation; other studies focus on applications involving classification and try to learn a representation from the signal to improve the classification result. Typically the goal is to define a class-specific sub-dictionary by using the signals of the corresponding class [8]. Previously, studies such as [13] suggested defining a common graph for all classes and learn class-specific graph transforms based on the signals in each class which play the role of sub-dictionaries.…”

Section: Introductionmentioning

confidence: 99%

“…Previously, studies such as [13] suggested defining a common graph for all classes and learn class-specific graph transforms based on the signals in each class which play the role of sub-dictionaries. The lack of discrimination between the class-specific graph transforms, which comes from using only one graph for all classes, may reduce the classification performance [8]. The distinctive goal here is that of discriminative transform learning, to effectively tackle a classification problem involving network data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

EEG-Based Emotion Classification Using Graph Signal Processing

Saboksayr

Mateos

Çetin

2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

The key role of emotions in human life is undeniable. The question of whether there exists a brain pattern associated with a specific emotion is the theme of many affective neuroscience studies. In this work, we bring to bear graph signal processing (GSP) techniques to tackle the problem of automatic emotion recognition using brain signals. GSP is an extension of classical signal processing methods to complex networks where there exists an inherent relation graph. With the help of GSP, we propose a new framework for learning class-specific discriminative graphs. To that end, firstly we assume for each class of observations there exists a latent underlying graph representation. Secondly, we consider the observations are smooth on their corresponding class-specific sough graph while they are non-smooth on other classes' graphs. The learned classspecific graph-based representations can act as sub-dictionaries and be utilized for the task of emotion classification. Applying the proposed method on an electroencephalogram (EEG) emotion recognition dataset indicates the superiority of our framework over other state-of-the-art methods.

show abstract

Disc-GLasso: Discriminative graph learning with sparsity regularization

Cited by 13 publications

References 9 publications

Deep neural networks on graph signals for brain imaging analysis

Deep neural networks on graph signals for brain imaging analysis

A Laplacian approach to $$\ell _1$$-norm minimization

EEG-Based Emotion Classification Using Graph Signal Processing

Contact Info

Product

Resources

About