Tensor Decomposition for Signal Processing and Machine Learning

被引:1070
作者
Sidiropoulos, Nicholas D. [1 ]
De Lathauwer, Lieven [2 ]
Fu, Xiao [1 ]
Huang, Kejun [1 ]
Papalexakis, Evangelos E. [3 ]
Faloutsos, Christos [4 ]
机构
[1] Univ Minnesota, Dept Elect & Comp Engn, Minneapolis, MN 55455 USA
[2] Katholieke Univ Leuven, B-3000 Leuven, Belgium
[3] Univ Calif Riverside, Dept Comp Sci, Riverside, CA 92521 USA
[4] Carnegie Mellon Univ, Dept Comp Sci, Pittsburgh, PA 15213 USA
基金
美国国家科学基金会; 欧洲研究理事会;
关键词
Tensor decomposition; tensor factorization; rank; canonical polyadic decomposition (CPD); parallel factor analysis (PARAFAC); Tucker model; higher-order singular value decomposition (HOSVD); multilinear singular value decomposition (MLSVD); uniqueness; NP-hard problems; alternating optimization; alternating direction method of multipliers; gradient descent; Gauss-Newton; stochastic gradient; Cramer-Rao bound; communications; source separation; harmonic retrieval; speech separation; collaborative filtering; mixture modeling; topic modeling; classification; subspace learning; CANONICAL POLYADIC DECOMPOSITION; NONNEGATIVE MATRIX FACTORIZATION; CRAMER-RAO BOUNDS; UNIQUENESS CONDITIONS; PART II; MULTILINEAR DECOMPOSITION; RANK APPROXIMATION; BLIND SEPARATION; MULTIWAY DATA; 3-WAY ARRAYS;
D O I
10.1109/TSP.2017.2690524
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
Tensors or multiway arrays are functions of three or more indices (i, j, k,...)-similar to matrices (two-way arrays), which are functions of two indices (r, c) for (row, column). Tensors have a rich history, stretching over almost a century, and touching upon numerous disciplines; but they have only recently become ubiquitous in signal and data analytics at the confluence of signal processing, statistics, data mining, and machine learning. This overview article aims to provide a good starting point for researchers and practitioners interested in learning about and working with tensors. As such, it focuses on fundamentals and motivation (using various application examples), aiming to strike an appropriate balance of breadth and depth that will enable someone having taken first graduate courses in matrix algebra and probability to get started doing research and/or developing tensor algorithms and software. Some background in applied optimization is useful but not strictly required. The material covered includes tensor rank and rank decomposition; basic tensor factorization models and their relationships and properties (including fairly good coverage of identifiability); broad coverage of algorithms ranging from alternating optimization to stochastic gradient; statistical performance analysis; and applications ranging from source separation to collaborative filtering, mixture and topic modeling, classification, and multilinear subspace learning.
引用
收藏
页码:3551 / 3582
页数:32
相关论文
共 159 条
  • [71] Putting Nonnegative Matrix Factorization to the Test [A tutorial derivation of pertinent Cramer-Rao bounds and performance benchmarking]
    Huang, Kejun
    Sidiropoulos, Nicholas D.
    [J]. IEEE SIGNAL PROCESSING MAGAZINE, 2014, 31 (03) : 76 - 86
  • [72] Non-Negative Matrix Factorization Revisited: Uniqueness and Algorithm for Symmetric Decomposition
    Huang, Kejun
    Sidiropoulos, Nicholas D.
    Swami, Ananthram
    [J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2014, 62 (01) : 211 - 224
  • [73] BEST LOW MULTILINEAR RANK APPROXIMATION OF HIGHER-ORDER TENSORS, BASED ON THE RIEMANNIAN TRUST-REGION SCHEME
    Ishteva, Mariya
    Absil, P. -A.
    Van Huffel, Sabine
    De Lathauwer, Lieven
    [J]. SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 2011, 32 (01) : 115 - 135
  • [74] Kruskal's permutation, lemma and the identification of CANDECOMP/PARAFAC and bilinear models with constant modulus constraints
    Jiang, T
    Sidiropoulos, ND
    [J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2004, 52 (09) : 2625 - 2636
  • [75] Almost-sure identifiability of multidimensional harmonic retrieval
    Jiang, T
    Sidiropoulos, ND
    ten Berge, JMF
    [J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2001, 49 (09) : 1849 - 1859
  • [76] Kang U., 2012, Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '12, P316, DOI DOI 10.1145/2339530.2339583
  • [77] Kofidis Eleftherios., 2001, STRUCTURED MATRICES, V280, P103
  • [78] Kolda T., 2006, P SDM WORKSH LINK AN
  • [79] Tensor Decompositions and Applications
    Kolda, Tamara G.
    Bader, Brett W.
    [J]. SIAM REVIEW, 2009, 51 (03) : 455 - 500
  • [80] Higher-order web link analysis using multilinear algebra
    Kolda, TG
    Bader, BW
    Kenny, JP
    [J]. FIFTH IEEE INTERNATIONAL CONFERENCE ON DATA MINING, PROCEEDINGS, 2005, : 242 - 249