Sparse nonnegative tensor decomposition using proximal algorithm and inexact block coordinate descent scheme
Wang, D., Chang, Z., & Cong, F. (2021). Sparse nonnegative tensor decomposition using proximal algorithm and inexact block coordinate descent scheme. Neural Computing and Applications, 33(24), 17369-17387. https://doi.org/10.1007/s00521-021-06325-8
Published inNeural Computing and Applications
DisciplineSecure Communications Engineering and Signal ProcessingOhjelmisto- ja tietoliikennetekniikkaTietotekniikkaSecure Communications Engineering and Signal ProcessingSoftware and Communications EngineeringMathematical Information Technology
© The Author(s) 2021
Nonnegative tensor decomposition is a versatile tool for multiway data analysis, by which the extracted components are nonnegative and usually sparse. Nevertheless, the sparsity is only a side effect and cannot be explicitly controlled without additional regularization. In this paper, we investigated the nonnegative CANDECOMP/PARAFAC (NCP) decomposition with the sparse regularization item using l1-norm (sparse NCP). When high sparsity is imposed, the factor matrices will contain more zero components and will not be of full column rank. Thus, the sparse NCP is prone to rank deficiency, and the algorithms of sparse NCP may not converge. In this paper, we proposed a novel model of sparse NCP with the proximal algorithm. The subproblems in the new model are strongly convex in the block coordinate descent (BCD) framework. Therefore, the new sparse NCP provides a full column rank condition and guarantees to converge to a stationary point. In addition, we proposed an inexact BCD scheme for sparse NCP, where each subproblem is updated multiple times to speed up the computation. In order to prove the effectiveness and efficiency of the sparse NCP with the proximal algorithm, we employed two optimization algorithms to solve the model, including inexact alternating nonnegative quadratic programming and inexact hierarchical alternating least squares. We evaluated the proposed sparse NCP methods by experiments on synthetic, real-world, small-scale, and large-scale tensor data. The experimental results demonstrate that our proposed algorithms can efficiently impose sparsity on factor matrices, extract meaningful sparse components, and outperform state-of-the-art methods. ...
Publication in research information system
MetadataShow full item record
Additional information about fundingOpen access funding provided by University of Jyväskylä (JYU). This work was supported by National Natural Science Foundation of China (Grant No.91748105), National Foundation in China (No. JCKY2019110B009 & 2020-JCJQ-JJ-252), the Fundamental Research Funds for the Central Universities [DUT20LAB303 & DUT20LAB308] in Dalian University of Technology in China, and the scholarship from China Scholarship Council (No. 201600090043). ...
Showing items with similar title or keywords.
Wang, Deqing; Cong, Fengyu; Ristaniemi, Tapani (IEEE, 2019)Tensor decomposition is a powerful tool for analyzing multiway data. Nowadays, with the fast development of multisensor technology, more and more data appear in higherorder (order > 4) and nonnegative form. However, the ...
Wang, Deqing; Zhu, Yongjie; Ristaniemi, Tapani; Cong, Fengyu (Elsevier BV, 2018)Background Preprocessed Event-related potential (ERP) data are usually organized in multi-way tensor, in which tensor decomposition serves as a powerful tool for data processing. Due to the limitation of computation burden ...
Wang, Xiulin; Ristaniemi, Tapani; Cong, Fengyu (IEEE, 2019)Real-world data exhibiting high order/dimensionality and various couplings are linked to each other since they share some common characteristics. Coupled tensor decomposition has become a popular technique for group ...
Low-rank approximation based non-negative multi-way array decomposition on event-related potentials Cong, Fengyu; Zhou, Guoxu; Astikainen, Piia; Zhao, Qibin; Wu, Qiang; Nandi, Asoke; Hietanen, Jari K.; Ristaniemi, Tapani; Cichocki, Andrzej (World Scientific, 2014)Non-negative tensor factorization (NTF) has been successfully applied to analyze event-related potentials (ERPs), and shown superiority in terms of capturing multi-domain features. However, the time-frequency representation ...
Zhu, Yongjie; Liu, Jia; Ye, Chaoxiong; Mathiak, Klaus; Astikainen, Piia; Ristaniemi, Tapani; Cong, Fengyu (Elsevier, 2020)Efficient neuronal communication between brain regions through oscillatory synchronization at certain frequencies is necessary for cognition. Such synchronized networks are transient and dynamic, established on the timescale ...