On the complexity of computing error bounds
From MaRDI portal
The authors study the cost of estimating the norm of an inverse matrix. They conjecture that finding even a coarse error bound is as costly as that of matrix inversion itself. Conversely, any fast condition estimate must sometimes mis-estimate by a sizable amount, specified quantitatively in the paper.
Recommendations
Cited in
(16)- On Koetter's algorithm and the computation of error values
- Probabilistic condition number estimates for real polynomial systems. I: A broader family of distributions
- On a generalized notion of mistake bounds
- Highly accurate verified error bounds for Krylov type linear system solvers
- Computing unrestricted synopses under maximum error bound
- The average condition number of most tensor rank decomposition problems is infinite
- Complexity estimates depending on condition and round-off error
- Resource Bounded Frequency Computations with Three Errors
- scientific article; zbMATH DE number 1206068 (Why is no real title available?)
- Tight and Rigorous Error Bounds for Basic Building Blocks of Double-Word Arithmetic
- scientific article; zbMATH DE number 2163480 (Why is no real title available?)
- New upper bounds on error exponents
- Subspace Iteration Randomization and Singular Value Problems
- Efficient and accurate computation of upper bounds of approximation errors
- On certain computable tests and componentwise error bounds
- Finding the orthogonal projection of a point onto an affine subspace
This page was built for publication: On the complexity of computing error bounds
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5944040)
