Efficient Nonnegative Tensor Decomposition Using Alternating Direction Proximal Method of Multipliers

被引:2
作者
Wang, Deqing [1 ,2 ,3 ]
Hu, Guoqiang [4 ]
机构
[1] Chinese Acad Sci, Shenyang Inst Automat, State Key Lab Robot, Shenyang 110016, Peoples R China
[2] Chinese Acad Sci, Inst Robot & Intelligent Mfg, Shenyang 110169, Peoples R China
[3] Key Lab Marine Robot, Shenyang 110169, Liaoning, Peoples R China
[4] Dalian Maritime Univ, Coll Artificial Intelligence, Dalian 116026, Peoples R China
关键词
Nonnegative CANDECOMP/PARAFAC; Alternating direction proximal method of multipliers; Tensor decomposition; Proximal algorithm; Sparse regularization; FACTORIZATION; OPTIMIZATION; ORGANIZATION; ALGORITHMS; MATRIX;
D O I
10.23919/cje.2023.00.035
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Nonnegative CANDECOMP/PARAFAC (NCP) tensor decomposition is a powerful tool for multiway signal processing. The alternating direction method of multipliers (ADMM) optimization algorithm has become increasingly popular for solving tensor decomposition problems in the block coordinate descent framework. However, the ADMM-based NCP algorithm suffers from rank deficiency and slow convergence for some large-scale and highly sparse tensor data. The proximal algorithm is preferred to enhance optimization algorithms and improve convergence properties. In this study, we propose a novel NCP algorithm using the alternating direction proximal method of multipliers (ADPMM) that consists of the proximal algorithm. The proposed NCP algorithm can guarantee convergence and overcome the rank deficiency. Moreover, we implement the proposed NCP using an inexact scheme that alternatively optimizes the subproblems. Each subproblem is optimized by a finite number of inner iterations yielding fast computation speed. Our NCP algorithm is a hybrid of alternating optimization and ADPMM and is named A2DPMM. 2 DPMM. The experimental results on synthetic and real-world tensors demonstrate the effectiveness and efficiency of our proposed algorithm.
引用
收藏
页码:1308 / 1316
页数:9
相关论文
共 34 条
  • [1] Exploring the reproducibility of functional connectivity alterations in Parkinson's disease
    Badea, Liviu
    Onu, Mihaela
    Wu, Tao
    Roceanu, Adina
    Bajenaru, Ovidiu
    [J]. PLOS ONE, 2017, 12 (11):
  • [2] BADER B. W., 2021, TENSOR TOOLBOX MATLA
  • [3] Bertsekas D., 2016, NONLINEAR PROGRAMMIN
  • [4] Distributed optimization and statistical learning via the alternating direction method of multipliers
    Boyd S.
    Parikh N.
    Chu E.
    Peleato B.
    Eckstein J.
    [J]. Foundations and Trends in Machine Learning, 2010, 3 (01): : 1 - 122
  • [5] The organization of the human striatum estimated by intrinsic functional connectivity
    Choi, Eun Young
    Yeo, B. T. Thomas
    Buckner, Randy L.
    [J]. JOURNAL OF NEUROPHYSIOLOGY, 2012, 108 (08) : 2242 - 2263
  • [6] Nonnegative matrix and tensor factorization
    Cichocki, Andrzej
    Zdunek, Rafal
    Amari, Shun-Ichi
    [J]. IEEE SIGNAL PROCESSING MAGAZINE, 2008, 25 (01) : 142 - 145
  • [7] Tensor Decompositions for Signal Processing Applications
    Cichocki, Andrzej
    Mandic, Danilo P.
    Anh Huy Phan
    Caiafa, Cesar F.
    Zhou, Guoxu
    Zhao, Qibin
    De Lathauwer, Lieven
    [J]. IEEE SIGNAL PROCESSING MAGAZINE, 2015, 32 (02) : 145 - 163
  • [8] Fast Local Algorithms for Large Scale Nonnegative Matrix and Tensor Factorizations
    Cichocki, Andrzej
    Phan, Anh-Huy
    [J]. IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES, 2009, E92A (03) : 708 - 721
  • [9] Frequency specific co-activation pattern analysis via sparse nonnegative tensor decomposition
    Hu, Guoqiang
    Wang, Deqing
    Luo, Siwen
    Hao, Yuxing
    Nickerson, Lisa D.
    Cong, Fengyu
    [J]. JOURNAL OF NEUROSCIENCE METHODS, 2021, 362
  • [10] A Flexible and Efficient Algorithmic Framework for Constrained Matrix and Tensor Factorization
    Huang, Kejun
    Sidiropoulos, Nicholas D.
    Liavas, Athanasios P.
    [J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2016, 64 (19) : 5052 - 5065