Krylov Subspace Methods for Tensor Computations
نویسنده
چکیده
A couple of generalizations of matrix Krylov subspace methods to tensors are presented. It is shown that a particular variant can be interpreted as a Krylov factorization of the tensor. A generalization to tensors of the Krylov-Schur method for computing matrix eigenvalues is proposed. The methods are intended for the computation of lowrank approximations of large and sparse tensors. A few numerical experiments are reported.
منابع مشابه
Krylov-type methods for tensor computations I
Several Krylov-type procedures are introduced that generalize matrix Krylov methods for tensor computations. They are denoted minimal Krylov recursion, maximal Krylov recursion, and contracted tensor product Krylov recursion. It is proved that, for a given tensor A with multilinear rank-(p, q, r), the minimal Krylov recursion extracts the correct subspaces associated to the tensor in p+ q+r num...
متن کاملTensor-Krylov methods for large nonlinear equations
In this paper, we describe tensor methods for large systems of nonlinear equations based on Krylov subspace techniques for approximately solving the linear systems that are required in each tensor iteration. We refer to a method in this class as a tensor-Krylov algorithm. We describe comparative testing for a tensor-Krylov implementation versus an analogous implementation based on a Newton-Kryl...
متن کاملOn the Numerical Stability Analysis of Pipelined Krylov Subspace Methods
Algebraic solvers based on preconditioned Krylov subspace methods are among the most powerful tools for large scale numerical computations in applied mathematics, sciences, technology, as well as in emerging applications in social sciences. The study of mathematical properties of Krylov subspace methods, in both the cases of exact and inexact computations, is a very active area of research and ...
متن کاملWedderburn rank reduction and Krylov subspace method for tensor approximation. Part 1: Tucker case
We propose algorithms for Tucker approximation of 3-tensor, that use it only through tensor-by-vector-by-vector multiplication subroutine. In matrix case, Krylov methods are methods of choice to approximate dominant column and row subspace of sparse or structured matrix given through matrix-by-vector subroutine. However, direct generalization to tensor case, proposed recently by Eldén and Savas...
متن کاملThe Numerical Stability Analysis of Pipelined Conjugate Gradient Methods: Historical Context and Methodology
Algebraic solvers based on preconditioned Krylov subspace methods are among the most powerful tools for large scale numerical computations in applied mathematics, sciences, technology, as well as in emerging applications in social sciences. As the name suggests, Krylov subspace methods can be viewed as a sequence of projections onto nested subspaces of increasing dimension. They are therefore b...
متن کامل