Increasing Stability of EEG Components Extraction Using Sparsity Regularized Tensor Decomposition

Deqing Wang; Xiaoyu Wang; Yongjie Zhu; Petri Toiviainen; Minna  Huotilainen; Tapani Ristaniemi; Fengyu Cong

Increasing Stability of EEG Components Extraction Using Sparsity Regularized Tensor Decomposition

Deqing Wang, Xiaoyu Wang, Yongjie Zhu, Petri Toiviainen, Minna Huotilainen, Tapani Ristaniemi, Fengyu Cong

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

Abstract

Tensor decomposition has been widely employed for EEG signal processing in recent years. Constrained and regularized tensor decomposition often attains more meaningful and interpretable results. In this study, we applied sparse nonnegative CANDECOMP/PARAFAC tensor decomposition to ongoing EEG data under naturalistic music stimulus. Interesting temporal, spectral and spatial components highly related with music features were extracted. We explored the ongoing EEG decomposition results and properties in a wide range of sparsity levels, and proposed a paradigm to select reasonable sparsity regularization parameters. The stability of interesting components extraction from fourteen subjects’ data was deeply analyzed. Our results demonstrate that appropriate sparsity regularization can increase the stability of interesting components significantly and remove weak components at the same …

Original language	English
Title of host publication	Advances in Neural Networks – ISNN 2018 : 15th International Symposium on Neural Networks, ISNN 2018, Minsk, Belarus, June 25–28, 2018, Proceedings
Editors	Tingwen Huang, Jiancheng Lv, Changyin Sun, Alexander V. Tuzikov
Number of pages	11
Publisher	Springer
Publication date	2018
Pages	789-799
ISBN (Print)	978-3-319-92536-3
ISBN (Electronic)	978-3-319-92537-0
Publication status	Published - 2018
MoE publication type	A4 Article in conference proceedings
Event	15th International Symposium on Neural Networks - Minsk, Belarus Duration: 25 Jun 2018 → 28 Jun 2018

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Number	10878
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Fields of Science

516 Educational sciences
3112 Neurosciences
3111 Biomedicine

Access to Document

https://link.springer.com/chapter/10.1007/978-3-319-92537-0_89#citeas

Cite this

Wang, D., Wang, X., Zhu, Y., Toiviainen, P., Huotilainen, M., Ristaniemi, T., & Cong, F. (2018). Increasing Stability of EEG Components Extraction Using Sparsity Regularized Tensor Decomposition. In T. Huang, J. Lv, C. Sun, & A. V. Tuzikov (Eds.), Advances in Neural Networks – ISNN 2018 : 15th International Symposium on Neural Networks, ISNN 2018, Minsk, Belarus, June 25–28, 2018, Proceedings (pp. 789-799). (Lecture Notes in Computer Science ; No. 10878). Springer. https://link.springer.com/chapter/10.1007/978-3-319-92537-0_89#citeas

Wang, Deqing ; Wang, Xiaoyu ; Zhu, Yongjie et al. / Increasing Stability of EEG Components Extraction Using Sparsity Regularized Tensor Decomposition. Advances in Neural Networks – ISNN 2018 : 15th International Symposium on Neural Networks, ISNN 2018, Minsk, Belarus, June 25–28, 2018, Proceedings . editor / Tingwen Huang ; Jiancheng Lv ; Changyin Sun ; Alexander V. Tuzikov. Springer, 2018. pp. 789-799 (Lecture Notes in Computer Science ; 10878).

@inproceedings{b3919b0dd5fc48cfa14225ebc75b3c7c,

title = "Increasing Stability of EEG Components Extraction Using Sparsity Regularized Tensor Decomposition",

abstract = "Tensor decomposition has been widely employed for EEG signal processing in recent years. Constrained and regularized tensor decomposition often attains more meaningful and interpretable results. In this study, we applied sparse nonnegative CANDECOMP/PARAFAC tensor decomposition to ongoing EEG data under naturalistic music stimulus. Interesting temporal, spectral and spatial components highly related with music features were extracted. We explored the ongoing EEG decomposition results and properties in a wide range of sparsity levels, and proposed a paradigm to select reasonable sparsity regularization parameters. The stability of interesting components extraction from fourteen subjects{\textquoteright} data was deeply analyzed. Our results demonstrate that appropriate sparsity regularization can increase the stability of interesting components significantly and remove weak components at the same …",

keywords = "516 Educational sciences, 3112 Neurosciences, 3111 Biomedicine",

author = "Deqing Wang and Xiaoyu Wang and Yongjie Zhu and Petri Toiviainen and Minna Huotilainen and Tapani Ristaniemi and Fengyu Cong",

year = "2018",

language = "English",

isbn = "978-3-319-92536-3",

series = "Lecture Notes in Computer Science ",

publisher = "Springer",

number = "10878",

pages = "789--799",

editor = "Huang, {Tingwen } and Lv, {Jiancheng } and Sun, {Changyin } and Tuzikov, {Alexander V.}",

booktitle = "Advances in Neural Networks – ISNN 2018",

address = "United Kingdom",

note = "15th International Symposium on Neural Networks ; Conference date: 25-06-2018 Through 28-06-2018",

}

Wang, D, Wang, X, Zhu, Y, Toiviainen, P, Huotilainen, M, Ristaniemi, T & Cong, F 2018, Increasing Stability of EEG Components Extraction Using Sparsity Regularized Tensor Decomposition. in T Huang, J Lv, C Sun & AV Tuzikov (eds), Advances in Neural Networks – ISNN 2018 : 15th International Symposium on Neural Networks, ISNN 2018, Minsk, Belarus, June 25–28, 2018, Proceedings . Lecture Notes in Computer Science , no. 10878, Springer, pp. 789-799, 15th International Symposium on Neural Networks, Minsk, Belarus, 25/06/2018. <https://link.springer.com/chapter/10.1007/978-3-319-92537-0_89#citeas>

Increasing Stability of EEG Components Extraction Using Sparsity Regularized Tensor Decomposition. / Wang, Deqing; Wang, Xiaoyu; Zhu, Yongjie et al.

Advances in Neural Networks – ISNN 2018 : 15th International Symposium on Neural Networks, ISNN 2018, Minsk, Belarus, June 25–28, 2018, Proceedings . ed. / Tingwen Huang; Jiancheng Lv; Changyin Sun; Alexander V. Tuzikov. Springer, 2018. p. 789-799 (Lecture Notes in Computer Science ; No. 10878).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Increasing Stability of EEG Components Extraction Using Sparsity Regularized Tensor Decomposition

AU - Wang, Deqing

AU - Wang, Xiaoyu

AU - Zhu, Yongjie

AU - Toiviainen, Petri

AU - Huotilainen, Minna

AU - Ristaniemi, Tapani

AU - Cong, Fengyu

PY - 2018

Y1 - 2018

N2 - Tensor decomposition has been widely employed for EEG signal processing in recent years. Constrained and regularized tensor decomposition often attains more meaningful and interpretable results. In this study, we applied sparse nonnegative CANDECOMP/PARAFAC tensor decomposition to ongoing EEG data under naturalistic music stimulus. Interesting temporal, spectral and spatial components highly related with music features were extracted. We explored the ongoing EEG decomposition results and properties in a wide range of sparsity levels, and proposed a paradigm to select reasonable sparsity regularization parameters. The stability of interesting components extraction from fourteen subjects’ data was deeply analyzed. Our results demonstrate that appropriate sparsity regularization can increase the stability of interesting components significantly and remove weak components at the same …

AB - Tensor decomposition has been widely employed for EEG signal processing in recent years. Constrained and regularized tensor decomposition often attains more meaningful and interpretable results. In this study, we applied sparse nonnegative CANDECOMP/PARAFAC tensor decomposition to ongoing EEG data under naturalistic music stimulus. Interesting temporal, spectral and spatial components highly related with music features were extracted. We explored the ongoing EEG decomposition results and properties in a wide range of sparsity levels, and proposed a paradigm to select reasonable sparsity regularization parameters. The stability of interesting components extraction from fourteen subjects’ data was deeply analyzed. Our results demonstrate that appropriate sparsity regularization can increase the stability of interesting components significantly and remove weak components at the same …

KW - 516 Educational sciences

KW - 3112 Neurosciences

KW - 3111 Biomedicine

M3 - Conference contribution

SN - 978-3-319-92536-3

T3 - Lecture Notes in Computer Science

SP - 789

EP - 799

BT - Advances in Neural Networks – ISNN 2018

A2 - Huang, Tingwen

A2 - Lv, Jiancheng

A2 - Sun, Changyin

A2 - Tuzikov, Alexander V.

PB - Springer

T2 - 15th International Symposium on Neural Networks

Y2 - 25 June 2018 through 28 June 2018

ER -

Wang D, Wang X, Zhu Y, Toiviainen P, Huotilainen M, Ristaniemi T et al. Increasing Stability of EEG Components Extraction Using Sparsity Regularized Tensor Decomposition. In Huang T, Lv J, Sun C, Tuzikov AV, editors, Advances in Neural Networks – ISNN 2018 : 15th International Symposium on Neural Networks, ISNN 2018, Minsk, Belarus, June 25–28, 2018, Proceedings . Springer. 2018. p. 789-799. (Lecture Notes in Computer Science ; 10878).