Neural networks for computing best rank-one approximations of tensors and its applications

Maolin Che, Andrzej Cichocki, Yimin Wei

    Research output: Contribution to journalArticlepeer-review

    28 Citations (Scopus)

    Abstract

    This paper presents the neural dynamical network to compute a best rank-one approximation of a real-valued tensor. We implement the neural network model by the ordinary differential equations (ODE), which is a class of continuous-time recurrent neural network. Several new properties of solutions for the neural network are established. We prove that the locally asymptotic stability of solutions for ODE by constructive an appropriate Lyapunov function under mild conditions. Furthermore, we also discuss how to use the proposed neural networks for solving the tensor eigenvalue problem including the tensor H-eigenvalue problem, the tensor Z-eigenvalue problem, and the generalized eigenvalue problem with symmetric-definite tensor pairs. Finally, we generalize the proposed neural networks to the computation of the restricted singular values and the associated restricted singular vectors of real-valued tensors. We illustrate and validate theoretical results via numerical simulations.

    Original languageEnglish
    Pages (from-to)114-133
    Number of pages20
    JournalNeurocomputing
    Volume267
    DOIs
    Publication statusPublished - 6 Dec 2017

    Keywords

    • Best rank-one approximation
    • Generalized tensor eigenpair
    • H-eigenpair
    • Local optimal rank-one approximation
    • Lyapunov function
    • Lyapunov stability theory
    • Neural network
    • Ordinary differential equations
    • Rank-one tensor
    • Restricted singular value
    • Restricted singular vector
    • Symmetric-definite tensor pair
    • The local maximal generalized eigenpair
    • The local minimal generalized eigenpair
    • Z-eigenpair

    Fingerprint

    Dive into the research topics of 'Neural networks for computing best rank-one approximations of tensors and its applications'. Together they form a unique fingerprint.

    Cite this