Low-Rank Updates of Matrix Functions II: Rational Krylov Methods
نویسندگان
چکیده
This work develops novel rational Krylov methods for updating a large-scale matrix function $f(A)$ when $A$ is subject to low-rank modifications. It extends our previous in this context on pol...
منابع مشابه
Rational Matrix Functions and Rank-1 Updates
Suppose f = p/q is a quotient of two polynomials and that p has degree rp and q has degree rq . Assume that f(A) and f(A+ uv ) are defined where A ∈ R, u ∈ R, and v ∈ R are given and set r = max{rp, rq}. We show how to compute f(A+uv ) in O(rn2) flops assuming that f(A) is available together with an appropriate factorization of the “denominator matrix” q(A). The central result can be interprete...
متن کاملLow-rank updates of matrix functions
Abstract. We consider the task of updating a matrix function f(A) when the matrix A ∈ Cn×n is subject to a low-rank modification. In other words, we aim at approximating f(A+D)− f(A) for a matrix D of rank k n. The approach proposed in this paper attains efficiency by projecting onto tensorized Krylov subspaces produced by matrix-vector multiplications with A and A∗. We prove the approximations...
متن کاملRational Krylov approximation of matrix functions: Numerical methods and optimal pole selection∗
Matrix functions are a central topic of linear algebra, and problems of their numerical approximation appear increasingly often in scientific computing. We review various rational Krylov methods for the computation of large-scale matrix functions. Emphasis is put on the rational Arnoldi method and variants thereof, namely, the extended Krylov subspace method and the shift-and-invert Arnoldi met...
متن کاملRational Krylov sequences and Orthogonal Rational Functions
In this paper we study the relationship between spectral decomposition, orthogonal rational functions and the rational Lanczos algorithm, based on a simple identity for rational Krylov sequences.
متن کاملLow-Rank Tensor Krylov Subspace Methods for Parametrized Linear Systems
We consider linear systems A(α)x(α) = b(α) depending on possibly many parameters α = (α1, . . . ,αp). Solving these systems simultaneously for a standard discretization of the parameter space would require a computational effort growing exponentially in the number of parameters. We show that this curse of dimensionality can be avoided for sufficiently smooth parameter dependencies. For this pur...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: SIAM Journal on Numerical Analysis
سال: 2021
ISSN: ['0036-1429', '1095-7170']
DOI: https://doi.org/10.1137/20m1362553