Abstract:The spectral transformation Lanczos method for the sparse symmetric definite generalized eigenvalue problem for matrices $A$ and $B$ is an iterative method that addresses the case of semidefinite or ill conditioned $B$ using a shifted and inverted formulation of the problem. This paper proposes the same approach for dense problems and shows that with a shift chosen in accordance with certain constraints, the algorithm can conditionally ensure that every computed shifted and inverted eigenvalue is close to the exact shifted and inverted eigenvalue of a pair of matrices close to $A$ and $B$. Under the same assumptions on the shift, the analysis of the algorithm for the shifted and inverted problem leads to useful error bounds for the original problem, including a bound that shows how a single shift that is of moderate size in a scaled sense can be chosen so that every computed generalized eigenvalue corresponds to a generalized eigenvalue of a pair of matrices close to $A$ and $B$. The computed generalized eigenvectors give a relative residual that depends on the distance between the corresponding generalized eigenvalue and the shift. If the shift is of moderate size, then relative residuals are small for generalized eigenvalues that are not much larger than the shift. Larger shifts give small relative residuals for generalized eigenvalues that are not much larger or smaller than the shift.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is about the numerical solution methods for the Symmetric Semidefinite Generalized Eigenvalue Problem, especially how to improve the stability and accuracy of the algorithm when the matrix $B$ is singular or ill - conditioned. Specifically, the paper mainly focuses on the generalized eigenvalue problem in the following form: \[Av = \lambda Bv, \quad v\neq0\] where $A$ and $B$ are symmetric matrices, and $B$ is a semi - positive definite or positive definite matrix. Such problems are very common in fields such as structural engineering and have a long history in the research literature of numerical linear algebra. ### Main problems and solutions 1. **Limitations of existing methods**: - When $B$ is positive definite, the commonly used method is the algorithm based on Cholesky decomposition. However, due to the possibly ill - conditioned Cholesky factors involved, the stability of this method is difficult to guarantee. - In practical applications, this algorithm usually gives a relatively small relative residual for large eigenvalues, but gives a relatively large relative residual for small eigenvalues. 2. **Proposed new method**: - The paper proposes a new algorithm by introducing the shifted and inverted formulation, that is, using symmetric factors to solve the system of the shifted matrix $A-\sigma B$. - By choosing an appropriate shift $\sigma$, the algorithm can be applied to error analysis and provide useful residual bounds. 3. **Key contributions of the algorithm**: - By choosing a suitable shift $\sigma$, the algorithm can ensure to a certain extent that each calculated shifted and inverted eigenvalue is close to the exact eigenvalue of a pair of matrices close to $A$ and $B$. - It provides useful residual bounds for the original problem, including one that shows how to choose a single moderately - sized shift to ensure that each calculated generalized eigenvalue corresponds to the generalized eigenvalue of a pair of matrices close to $A$ and $B$. 4. **Theoretical analysis and experimental verification**: - The paper derives the algorithm in detail and provides forward/backward stability results as well as related error bounds. - The accuracy and effectiveness of these error bounds are verified through numerical experiments. ### Summary The main purpose of the paper is to provide a more stable and efficient solution method for the Symmetric Semidefinite Generalized Eigenvalue Problem, especially when the $B$ matrix may be singular or ill - conditioned. By introducing the shift and inverse transformation techniques, the paper proposes a new algorithm and conducts a detailed theoretical analysis and experimental verification, proving its superiority in practical applications.

Spectral Transformation for the Dense Symmetric Semidefinite Generalized Eigenvalue Problem

Parallel Eigenvalue Calculation Based On Multiple Shift-Invert Lanczos And Contour Integral Based Spectral Projection Method

Estimation of spectral gaps for sparse symmetric matrices

The Convergence of Generalized Lanczos Methods for Large Unsymmetric Eigenproblems

A generalized skew-symmetric Lanczos bidiagonalization method for computing several extreme eigenpairs of a large skew-symmetric/symmetric positive definite matrix pair

A Skew-Symmetric Lanczos Bidiagonalization Method for Computing Several Extremal Eigenpairs of a Large Skew-Symmetric Matrix

A Refined Shift-And-Invert Arnoldi Algorithm for Large Unsymmetric Generalized Eigenproblems

Fast computation of spectral densities for generalized eigenvalue problems

Improved Spectral Density Estimation via Explicit and Implicit Deflation

Spectral Condition-Number Estimation of Large Sparse Matrices

A Communication-Avoiding Parallel Algorithm for the Symmetric Eigenvalue Problem

A Nesterov-style Accelerated Gradient Descent Algorithm for the Symmetric Eigenvalue Problem

A Shift Selection Strategy for Parallel Shift-Invert Spectrum Slicing in Symmetric Self-Consistent Eigenvalue Computation

Solving Dense Generalized Eigenproblems on Multi-threaded Architectures

A FEAST SVDsolver Based on Chebyshev–Jackson Series for Computing Partial Singular Triplets of Large Matrices

Localized spectrum slicing

Efficient Preconditioner Updates for Unsymmetric Shifted Linear Systems

Generalized Block Lanczos Methods for Large Unsymmetric Eigenproblems

Isira: Integrated Shift-Invert Residual Arnoldi Method for Graph Laplacian Matrices from Big Data

Spectrally Constrained Optimization

Gradient-type subspace iteration methods for the symmetric eigenvalue problem