Аннотация:
Пусть в сепарабельном гильбертовом или евклидовом пространстве задан линейный ограниченный самосопряженный оператор со спектром из [λmin,λmax] причем некоторый полуинтервал (ν,λmax] содержит лишь точечный спектр λmax≡λ1>⋯>λp, p<∞ и соответствующие собственные подпространства Uj конечномерны, U≡U1+⋯+Up. Через Θ(V;W) обозначим раствор подпространств V и W. В некотором «пробном» подпространстве ˜U определим приближения ˜λj, и ˜Uj. по методу Релея–Ритца. Тогда 0\le\lambda_j-\widetilde\lambda_j\le(\lambda_j-\lambda_{\min})\Theta^2(\widetilde U;U).
Погрешность для собственных подпространств:
\Theta^2(\widetilde U_1;U_1)\le(\lambda_1-\widetilde\lambda_1)/(\lambda_1-\lambda_2);
если для некоторого j\in[2,p] имеем \widetilde\lambda_{j-1}>\lambda_j и \widetilde\lambda_j>\lambda_{j+1}, то
\Theta^2(\widetilde U_j;U_j)\le(\widetilde\lambda_{j-1}-\widetilde\lambda_j)^{-1}(\lambda_j-\lambda_{j+1})^{-1}(\widetilde\lambda_{j-1}-\lambda_{j+1})\cdot(\lambda_j-\widetilde\lambda_j).
Библиогр. 12 назв.
Образец цитирования:
А. В. Князев, “Точные априорные оценки погрешности метода Релея–Ритца без предположений знакоопределенности или компактности”, Матем. заметки, 38:6 (1985), 900–907; Math. Notes, 38:6 (1985), 998–1002
Samuel M. Greene, Robert J. Webber, Timothy C. Berkelbach, Jonathan Weare, “Approximating Matrix Eigenvalues by Subspace Iteration with Repeated Random Sparsification”, SIAM J. Sci. Comput., 44:5 (2022), A3067
Robert J. Webber, Erik H. Thiede, Douglas Dow, Aaron R. Dinner, Jonathan Weare, “Error Bounds for Dynamical Spectral Estimation”, SIAM Journal on Mathematics of Data Science, 3:1 (2021), 225
Rebekka Gasser, Joscha Gedicke, Stefan Sauter, “Benchmark Computation of Eigenvalues with Large Defect for Non-Self-adjoint Elliptic Differential Operators”, SIAM J. Sci. Comput., 41:6 (2019), A3938
Zhongming Teng, Linzhang Lu, Ren-Cang Li, “Cluster-robust accuracy bounds for Ritz subspaces”, Linear Algebra and its Applications, 480 (2015), 11
Lei-Hong Zhang, Jungong Xue, Ren-Cang Li, “Rayleigh–Ritz Approximation For the Linear Response Eigenvalue Problem”, SIAM J. Matrix Anal. & Appl., 35:2 (2014), 765
Peizhen Zhu, Merico E. Argentati, Andrew V. Knyazev, “Bounds for the Rayleigh Quotient and the Spectrum of Self-Adjoint Operators”, SIAM J. Matrix Anal. & Appl., 34:1 (2013), 244
Daniele Boffi, Francesca Gardini, Lucia Gastaldi, Lecture Notes in Computational Science and Engineering, 85, Frontiers in Numerical Analysis - Durham 2010, 2011, 1
Andrew V. Knyazev, Merico E. Argentati, “Rayleigh–Ritz Majorization Error Bounds with Applications to FEM”, SIAM J. Matrix Anal. & Appl., 31:3 (2010), 1521
S. Sauter, “hp-Finite Elements for Elliptic Eigenvalue Problems: Error Estimates Which Are Explicit with Respect to \lambda, h, and p”, SIAM J. Numer. Anal., 48:1 (2010), 95
M. E. Argentati, A. V. Knyazev, C. C. Paige, I. Panayotov, “Bounds on Changes in Ritz Values for a Perturbed Invariant Subspace of a Hermitian Matrix”, SIAM J. Matrix Anal. & Appl., 30:2 (2008), 548
A. V. Knyazev, M. E. Argentati, I. Lashuk, E. E. Ovtchinnikov, “Block Locally Optimal Preconditioned Eigenvalue Xolvers (BLOPEX) in Hypre and PETSc”, SIAM J. Sci. Comput., 29:5 (2007), 2224
Andrew V. Knyazev, John E. Osborn, “New A Priori FEM Error Estimates for Eigenvalues”, SIAM J. Numer. Anal., 43:6 (2006), 2647
E. Ovtchinnikov, “Cluster robust error estimates for the Rayleigh–Ritz approximation I: Estimates for invariant subspaces”, Linear Algebra and its Applications, 415:1 (2006), 167
E. Ovtchinnikov, “Cluster robustness of preconditioned gradient subspace iteration eigensolvers”, Linear Algebra and its Applications, 415:1 (2006), 140
E. Ovtchinnikov, “Cluster robust error estimates for the Rayleigh–Ritz approximation II: Estimates for eigenvalues”, Linear Algebra and its Applications, 415:1 (2006), 188
Andrew V. Knyazev, Alexander L. Skorokhodov, “Preconditioned Gradient-Type Iterative Methods in a Subspace for Partial Generalized Symmetric Eigenvalue Problems”, SIAM J. Numer. Anal., 31:4 (1994), 1226
A.V. Knyazev, A.L. Shorokhodov, “On exact estimates of the convergence rate of the steepest ascent method in the symmetric eigenvalue problem”, Linear Algebra and its Applications, 154-156 (1991), 245