Abstract
Tensor decomposition is widely used in a variety of applications such as data mining, biomedical informatics, neuroscience, and signal processing. In this paper, we propose a deep non-negative tensor factorization (DNTF) model to learn intrinsic and hierarchical structures from multi-way data. The DNTF model takes the Tucker tensor decomposition as a building block to stack up a multi-layer structure. In such a way, we can gradually learn the more abstract structures in a higher layer. The benefit is that it helps to mine intrinsic correlations and hierarchical structures from multi-way data. The non-negative constraints allow for clustering interpretation of the extracted data-dependent components. The objective of DNTF is to minimize the total reconstruction loss resulting from using the core tensor in the highest layer and the mode matrices in each layer to reconstruct the data tensor. Then, a deep decomposition algorithm based on multiplicative update rules is proposed to solve the optimization problem. It first conducts layer-wise tensor factorization and then fine-tunes the weights of all layers to reduce the total reconstruction loss. The experimental results on biosignal sensor data demonstrate the effectiveness and robustness of the proposed approach.
Similar content being viewed by others
References
Kolda TG, Bader BW (2009) Tensor decompositions and applications. SIAM Rev 51(3):455–500
Zheng VW, Zheng Y, Xie X, Yang Q (2012) Towards mobile intelligence: learning from GPS history data for collaborative recommendation. Artif Intell 184–185:17–37
Kim Y, Cichocki A, Choi S (2008) Nonnegative tucker decomposition with alpha-divergence. In ICASSP, pages 1829–1832
Carroll JD, Chang J-J (1970) Analysis of individual differences in multidimensional scaling via an n-way generalization of Eckart-Young decomposition. Psychometrika 35:283–319
Harshman RA (1970) Foundations of the PARAFAC procedure: models and conditions for an explanatory multi-modal factor analysis. UCLA Working Papers in Phonetics 16:1–84
Tucker LR (1966) Some mathematical notes on three-mode factor analysis. Psychometrika 31:279–311
Shashua A, Hazan T (2005) Non-negative tensor factorization with applications to statistics and computer vision. In ICML, pages 792–799
Kim Y, Choi S (2007) Nonnegative tucker decomposition. In CVPR
Lee DD, Seung HS (2000) Algorithms for non-negative matrix factorization. In NIPS, pages 556–562
Pascual-Montano AD, Carazo JM, Kochi K, Lehmann D, Pascual-Marqui RD (2006) Nonsmooth nonnegative matrix factorization (nsNMF). IEEE Trans Pattern Anal Mach Intell 28(3):403–415
Ding CHQ, Li T, Peng W, Park H (2006) Orthogonal nonnegative matrix t-factorizations for clustering. In KDD, pages 126–135, 2006
Welling M, Weber M (2001) Positive tensor factorization. Pattern Recogn Lett 22(12):1255–1261
Yoshii K, Tomioka R, Mochihashi D, Goto M (2013) Infinite positive semidefinite tensor factorization for source separation of mixture signals. In ICML, pages 576–584
Mørup M, Hansen LK, Arnfred SM (2008) Algorithms for sparse nonnegative tucker decompositions. Neural Comput 20(8):2112–2131
Cichocki A, Zdunek R, Choi S, Plemmons RJ, Amari S (2007) Non-negative tensor factorization using alpha and beta divergences. In ICASSP, pages 1393–1396
Cichocki A, Zdunek R, Choi S, Plemmons RJ, Amari S (2007) Novel multi-layer non-negative tensor factorization with sparsity constraints. In ICANNGA, pages 271–280, 2007
Liu J, Liu J, Wonka P, Ye J (2012) Sparse non-negative tensor factorization using columnwise coordinate descent. Pattern Recogn 45(1):649–656
Trigeorgis G, Bousmalis K, Zafeiriou S, Schuller BW (2014) A deep semi-NMF model for learning hidden representations. In ICML, pages 1692–1700
Ding CHQ, Li T, Jordan MI (2010) Convex and semi-nonnegative matrix factorizations. IEEE Trans Pattern Anal Mach Intell 32(1):45–55
Yang P, He J (2015) Model multiple heterogeneity via hierarchical multi-latent space learning. In KDD, pages 1375–1384
Fan J, Cheng J (2018) Matrix completion by deep matrix factorization. Neural Netw 98:34–41
Zhang W, Zhang X, Wang H, Chen D (2019) A deep variational matrix factorization method for recommendation on large scale sparse dataset. Neurocomputing 334:206–218
Xu S, Zhang C, Zhang J (2020) Bayesian deep matrix factorization network for multiple images denoising. Neural Netw 123:420–428
Sun Z, Huang S, Jiang P, Hu P, Lu Z (2020) DTF: deep tensor factorization for predicting anticancer drug synergy. Bioinformatics 36(16):4483–4489
Jacob Schreiber JB, Timothy Durham, Noble WS (2020) Avocado: a multi-scale deep tensor factorization method learns a latent representation of the human epigenome. Genom Biol 21(8):1
Oseledets IV (2011) Tensor-train decomposition. SIAM J Sci Comput 33(5):2295–2317
Phinyomark PPA, Limsakul C (2012) Feature reduction and selection for EMG signal classification. Expert Syst Appl 39(8):7420–7431
Chen M, Zhou P (2016) A novel framework based on FastICA for high density surface EMG decomposition. IEEE Trans Neural Syst Rehabil Eng 24(1):117–127
Naik MGAAGR, Selvan SE, Nguyen HT (2016) Principal component analysis applied to surface electromyography: a comprehensive review. IEEE Access 4:4025–4037
Naik GR, Nguyen HT (2015) Nonnegative matrix factorization for the identification of EMG finger movements: evaluation using matrix analysis. IEEE J Biomed Health Inform 19(2):478–485
Rosati MKS, Agostini V, Balestra G (2017) Muscle activation patterns during gait: a hierarchical clustering analysis. Biomed Signal Process Control 31:463–469
B JL et al (2018) Limb position tolerant pattern recognition for myoelectric prosthesis control with adaptive sparse representations from extreme learning. IEEE Trans Biomed Eng 65(4):770–778
Simão M, Mendes N, Gibaru O, Neto P (2019) A review on electromyography decoding and pattern recognition for human-machine interaction. IEEE Access 7:39564–39582
Bengio Y (2009) Learning deep architectures for AI. Found. Trends Mach Learn 2(1):1–127
Hofmann T (1999) Probabilistic latent semantic analysis. In UAI, pages 289–296
Dua D, Graff C (2017) UCI machine learning repository
Chattopadhyay R, Jesunathadas M, Poston B, Santello M, Ye J, Panchanathan S (2012) A subject-independent method for automatically grading electromyographic features during a fatiguing contraction. IEEE Trans Biomed Eng 59(6):1749–1757
Acknowledgements
This work was supported by National Natural Science Foundation of China (Nos. 61976092 and 61473123), Guangdong Basic and Applied Basic Research Foundation (Nos. 2021A1515011317, 2017A030313370, and 2018A030313356), and Guangdong Key Area Research and Development Plan (No. 2020B1111120001).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Tan, Q., Yang, P. & Wen, G. Deep non-negative tensor factorization with multi-way EMG data. Neural Comput & Applic 34, 1307–1317 (2022). https://doi.org/10.1007/s00521-021-06474-w
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-021-06474-w