A1 Journal article (refereed)
An inexact alternating proximal gradient algorithm for nonnegative CP tensor decomposition (2021)

Wang, D., & Cong, F. (2021). An inexact alternating proximal gradient algorithm for nonnegative CP tensor decomposition. Science China: Technological Sciences, 64(9), 1893-1906. https://doi.org/10.1007/s11431-020-1840-4

JYU authors or editors

Publication details

All authors or editors: Wang, DeQing; Cong, FengYu

Journal or series: Science China: Technological Sciences

ISSN: 1674-7321

eISSN: 1869-1900

Publication year: 2021

Publication date: 20/07/2021

Volume: 64

Issue number: 9

Pages range: 1893-1906

Publisher: Springer Science and Business Media LLC

Publication country: China

Publication language: English

DOI: https://doi.org/10.1007/s11431-020-1840-4

Publication open access: Not open

Publication channel open access:


Nonnegative tensor decomposition has become increasingly important for multiway data analysis in recent years. The alternating proximal gradient (APG) is a popular optimization method for nonnegative tensor decomposition in the block coordinate descent framework. In this study, we propose an inexact version of the APG algorithm for nonnegative CANDECOMP/PARAFAC decomposition, wherein each factor matrix is updated by only finite inner iterations. We also propose a parameter warm-start method that can avoid the frequent parameter resetting of conventional APG methods and improve convergence performance. By experimental tests, we find that when the number of inner iterations is limited to around 10 to 20, the convergence speed is accelerated significantly without losing its low relative error. We evaluate our method on both synthetic and real-world tensors. The results demonstrate that the proposed inexact APG algorithm exhibits outstanding performance on both convergence speed and computational precision compared with existing popular algorithms.

Keywords: optimisation; mathematical optimisation; algorithms

Free keywords: tensor decomposition; nonnegative CANDECOMP/PARAFAC; block coordinate descent; alternating proximal gradient; inexact scheme

Contributing organizations

Ministry reporting: Yes

Reporting Year: 2021

JUFO rating: 1

Last updated on 2023-31-01 at 22:04