Uncertainty in the solution of linear operator equations
From MaRDI portal
The author introduces and explores a new measure of accuracy for the approximate solution of a linear operator equation \(Lx=y\) in normed spaces. Instead of the error bound \(\| x-\hat x\|\) he defines the uncertainty \(U_{\epsilon}(\hat x)=\sup\{\| z-\hat x\|:z\in S_{\epsilon}\}\) where \(S_{\epsilon}=\{x\in X:\| Lx-y\|\leq \epsilon\}\). This notion seems to be very useful especially in the case of ill-posed problems. Numerical examples demonstrate the applicability of the theory to some practical cases.
Recommendations
Cited in
(12)- What are the proper condition numbers of discretized ill-posed problems?
- Optimal residual algorithms for linear operator equations
- Regularization tools: A Matlab package for analysis and solution of discrete ill-posed problems
- On regularization and discretization control for the numerical solution of inverse problems in parabolic equations
- Approximations for ill-posed problems
- Modulus of continuity for conditionally stable ill-posed problems in Hilbert space
- scientific article; zbMATH DE number 3979818 (Why is no real title available?)
- scientific article; zbMATH DE number 4046974 (Why is no real title available?)
- A critique of numerical analysis
- Existence and uniqueness of solutions of uncertain linear systems
- Constructing analytical solutions with uncertainty for recurrent, matrix and integral equations
- Regularization of a Volterra integral equation by linear inequalities
This page was built for publication: Uncertainty in the solution of linear operator equations
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q797962)
