In this note we present a parameterized class of lower triangular matrices. The components of the eigenvectors grow rapidly and will exceed the representational range of any finite number system. The eigenvalues and the eigenvectors are well-conditioned with respect to componentwise relative perturbations of the matrix. This class of matrices is well suited for testing software for computing eigenvectors as these routines must be able to handle overflow successfully.
An Overflow Free Fixed-point Eigenvalue Decomposition Algorithm: Case Study of Dimensionality Reduction in Hyperspectral Images
Laboratoire d’Informatique de l’Ecole Polytechnique, CNRS, Université Paris-Saclay（巴黎综合理工学院计算机实验室，法国国家科学研究中心，巴黎萨克雷大学）；Space Application Center, Indian Space Research Organisation（空间应用中心，印度空间研究组织）；Techno India University（印度理工大学）
The condition number for eigenvalue computations is a well--studied quantity. But how small can we expect it to be? Namely, which is a perfectly conditioned matrix w.r.t. eigenvalue computations? In this note we answer this question with exact first order asymptotic.
Since numbers in the computer are represented with a fixed number of bits, loss of accuracy during calculation is unavoidable. At high precision where more bits (e.g. 64) are allocated to each number, round-off errors are typically small. On the other hand, calculating at lower precision, such as half (16 bits), has the advantage of being much faster. This research focuses on experimenting with arithmetic at different precision levels for large-scale inverse problems, which are represented by linear systems with ill-conditioned matrices. We modified the Conjugate Gradient Method for Least Squa…
There are some types of ill-conditioned algebraic equations that have difficulty in obtaining accurate roots and coefficients that must be expressed with a multiple precision floating-point number. When all their roots are simple, the problem solved via eigensolver (eigenvalue method) is well-conditioned if the corresponding companion matrix has its small condition number. However, directly solving them with Newton or simultaneous iteration methods (direct iterative method for short) should be considered as ill-conditioned because of increasing density of its root distribution. Although a grea…
This paper offers a review of numerical methods for computation of the eigenvalues of Hermitian matrices and the singular values of general and some classes of structured matrices. The focus is on the main principles behind the methods that guarantee high accuracy even in the cases that are ill-conditioned for the conventional methods. First, it is shown that a particular structure of the errors in a finite precision implementation of an algorithm allows for a much better measure of sensitivity and that computation with high accuracy is possible despite a large classical condition number. Such…
The numerical solution of singular eigenvalue problems is complicated by the fact that small perturbations of the coefficients may have an arbitrarily bad effect on eigenvalue accuracy. However, it has been known for a long time that such perturbations are exceptional and standard eigenvalue solvers, such as the QZ algorithm, tend to yield good accuracy despite the inevitable presence of roundoff error. Recently, Lotz and Noferini quantified this phenomenon by introducing the concept of $δ$-weak eigenvalue condition numbers. In this work, we consider singular quadratic eigenvalue problems and …
Modal and nonmodal analyses of fluid flows provide fundamental insight into the early stages of transition to turbulence. Eigenvalues of the dynamical generator govern temporal growth or decay of individual modes, while singular values of the frequency response operator quantify the amplification of disturbances for linearly stable flows. In this paper, we develop well-conditioned ultraspherical and spectral integration methods for frequency response analysis of channel flows of Newtonian and viscoelastic fluids. Even if a discretization method is well-conditioned, we demonstrate that calculat…
This report shows on real data that the direct methods such as LDL decomposition and Gaussian elimination for solving linear systems with ill-conditioned matrices provide inaccurate results due to divisions by very small numbers, which in turn results in peaking phenomena and large estimation errors. Richardson iteration provides accurate results without peaking phenomena since division by small numbers is absent in the Richardson approach. In addition, two preconditioners are considered and compared in the Richardson iteration: 1) the simplest and robust preconditioner based on the maximum ro…
We develop a new homotopy method for solving multiparameter eigenvalue problems (MEPs) called the fiber product homotopy method. For a $k$-parameter eigenvalue problem with matrices of sizes $n_1,\dots ,n_k = O(n)$, fiber product homotopy method requires deformation of $O(1)$ linear equations, while existing homotopy methods for MEPs require $O(n)$ nonlinear equations. We show that the fiber product homotopy method theoretically finds all eigenpairs of an MEP with probability one. It is especially well-suited for dimension-deficient singular MEPs, a weakness of all other existing methods, as t…
We consider the problem of extracting a few desired eigenpairs of the buckling eigenvalue problem $Kx = λK_Gx$, where $K$ is symmetric positive semi-definite, $K_G$ is symmetric indefinite, and the pencil $K - λK_G$ is singular, namely, $K$ and $K_G$ share a non-trivial common nullspace. Moreover, in practical buckling analysis of structures, bases for the nullspace of $K$ and the common nullspace of $K$ and $K_G$ are available. There are two open issues for developing an industrial strength shift-invert Lanczos method: (1) the shift-invert operator $(K - σK_G)^{-1}$ does not exist or is extre…
In many applications, linear systems arise where the coefficient matrix takes the special form ${\bf I} + {\bf K} + {\bf E}$, where ${\bf I}$ is the identity matrix of dimension $n$, ${\rm rank}({\bf K}) = p \ll n$, and $\|{\bf E}\| \leq ε< 1$. GMRES convergence rates for linear systems with coefficient matrices of the forms ${\bf I} + {\bf K}$ and ${\bf I} + {\bf E}$ are guaranteed by well-known theory, but only relatively weak convergence bounds specific to matrices of the form ${\bf I} + {\bf K} + {\bf E}$ currently exist. In this paper, we explore the convergence properties of linear syste…
An EigenValue Stabilization Technique for Immersed Boundary Finite Element Methods in Explicit Dynamics
Otto von Guericke University Magdeburg（马格德堡奥托·冯·格里克大学）；Hamburg University of Technology（汉堡工业大学）；Technische Universität Dresden（德累斯顿工业大学）；Technical University of Darmstadt（达姆施塔特工业大学）
We present adapted Zhang Neural Networks (AZNN) in which the parameter settings for the exponential decay constant $η$ and the length of the start-up phase of basic ZNN are adapted to the problem at hand. Specifically we study experiments with AZNN for time-varying square matrix factorizations as a product of time-varying symmetric matrices and for the time-varying matrix square roots problem. Differing from generally used small $η$ values and minimal start-up length phases in ZNN, we adapt the basic ZNN method to work with large or even gigantic $η$ settings and arbitrary length start-ups usi…
In the present note we consider a type of matrices stemming in the context of the numerical approximation of distributed order fractional differential equations (FDEs): from one side they could look standard, since they are, real, symmetric and positive definite. On the other hand they present specific difficulties which prevent the successful use of classical tools. In particular the associated matrix-sequence, with respect to the matrix-size, is ill-conditioned and it is such that a generating function does not exists, but we face the problem of dealing with a sequence of generating function…
