PRIMITIVITY AND HURWITZ PRIMITIVITY OF NONNEGATIVE MATRIX TUPLES: A UNIFIED APPROACH

被引:1
|
作者
Wu, Yaokun [1 ,2 ]
Zhu, Yinfeng [1 ,3 ]
机构
[1] Shanghai Jiao Tong Univ, Sch Math Sci, Shanghai 200240, Peoples R China
[2] Shanghai Jiao Tong Univ, MOE LSC, Shanghai 200240, Peoples R China
[3] Imperial Coll London, Dept Math, 180 Queens Gate, London SW7 2AZ, England
基金
中国国家自然科学基金;
关键词
automaton; vCern; 'y function; ergodic exponent; Hamiltonian walk; primitive expo-nent; stable relation; AUTOMATA; SETS; COVERINGS; GRAPHS; WORDS;
D O I
10.1137/22M1471535
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
For an m-tuple of nonnegative n\times n matrices (A1,...,Am), primitivity/Hurwitz primitivity means the existence of a positive product/Hurwitz product, respectively (all products are with repetitions permitted). The Hurwitz product with a Parikh vector a = (a1, . . . , am) \in Z\geqm0 is the sum of all products with ai multipliers Ai, i = 1, ... , m. Ergodicity/Hurwitz ergodicity means the existence of the corresponding product with a positive row. We give a unified proof for the Protasov-Vonyov characterization (2012) of primitive tuples of matrices without zero rows and columns and for the Protasov characterization (2013) of Hurwitz primitive tuples of matrices without zero rows. By establishing a connection with synchronizing automata, we, under the aforementioned conditions, find an O(n2m)-time algorithm to decide primitivity and an O(n3m2)-time algorithm to construct a Hurwitz primitive vector a of weight \summi=1 ai = O(n3). We also report results on ergodic and Hurwitz ergodic matrix tuples.
引用
收藏
页码:196 / 211
页数:16
相关论文
共 50 条
  • [21] A unified global convergence analysis of multiplicative update rules for nonnegative matrix factorization
    Norikazu Takahashi
    Jiro Katayama
    Masato Seki
    Jun’ichi Takeuchi
    Computational Optimization and Applications, 2018, 71 : 221 - 250
  • [22] A Unified Convergence Analysis of the Multiplicative Update Algorithm for Regularized Nonnegative Matrix Factorization
    Zhao, Renbo
    Tan, Vincent Y. F.
    IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2018, 66 (01) : 129 - 138
  • [23] UNMF: a unified nonnegative matrix factorization for multi-dimensional omics data
    Abe, Ko
    Shimamura, Teppei
    BRIEFINGS IN BIOINFORMATICS, 2023, 24 (05)
  • [24] A unified global convergence analysis of multiplicative update rules for nonnegative matrix factorization
    Takahashi, Norikazu
    Katayama, Jiro
    Seki, Masato
    Takeuchi, Jun'ichi
    COMPUTATIONAL OPTIMIZATION AND APPLICATIONS, 2018, 71 (01) : 221 - 250
  • [25] A Geometric Approach to Archetypal Analysis and Nonnegative Matrix Factorization
    Damle, Anil
    Sun, Yuekai
    TECHNOMETRICS, 2017, 59 (03) : 361 - 370
  • [26] A Dynamical System Approach for Continuous Nonnegative Matrix Factorization
    Melisew Tefera Belachew
    Nicoletta Del Buono
    Mediterranean Journal of Mathematics, 2017, 14
  • [27] A Quasi-Likelihood Approach to Nonnegative Matrix Factorization
    Devarajan, Karthik
    Cheung, Vincent C. K.
    NEURAL COMPUTATION, 2016, 28 (08) : 1663 - 1693
  • [28] A Provable Splitting Approach for Symmetric Nonnegative Matrix Factorization
    Li, Xiao
    Zhu, Zhihui
    Li, Qiuwei
    Liu, Kai
    IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2023, 35 (03) : 2206 - 2219
  • [29] A Collaborative Neurodynamic Approach to Symmetric Nonnegative Matrix Factorization
    Che, Hangjun
    Wang, Jun
    NEURAL INFORMATION PROCESSING (ICONIP 2018), PT II, 2018, 11302 : 453 - 462
  • [30] A Dynamical System Approach for Continuous Nonnegative Matrix Factorization
    Belachew, Melisew Tefera
    Del Buono, Nicoletta
    MEDITERRANEAN JOURNAL OF MATHEMATICS, 2017, 14 (01)