LSMR: An iterative algorithm for sparse least-squares problems

@article{Fong2010LSMRAI,
  title={LSMR: An iterative algorithm for sparse least-squares problems},
  author={D. C. Fong and M. Saunders},
  journal={ArXiv},
  year={2010},
  volume={abs/1006.0758}
}
An iterative method LSMR is presented for solving linear systems Ax = b and leastsquares problems min ‖Ax−b‖2, with A being sparse or a fast linear operator. LSMR is based on the Golub-Kahan bidiagonalization process. It is analytically equivalent to the MINRES method applied to the normal equation ATAx = ATb, so that the quantities ‖Ark‖ are monotonically decreasing (where rk = b−Axk is the residual for the current iterate xk). We observe in practice that ‖rk‖ also decreases monotonically, so… Expand
LSLQ: An Iterative Method for Linear Least-Squares with an Error Minimization Property
TLDR
An iterative method named LSLQ for solving linear least-squares problems of any shape, formally equivalent to SYMMLQ applied to the normal equations, so that the current estimate's Euclidean norm increases monotonically, while the associated error norm decreases monotonally. Expand
The block LSMR algorithm for solving linear systems with multiple right-hand sides
LSMR (Least Squares Minimal Residual) is an iterative method for the solution of the linear system of equations and leastsquares problems. This paper presents a block version of the LSMR algorithmExpand
LSRN: A Parallel Iterative Solver for Strongly Over- or Underdetermined Systems
TLDR
A parallel iterative least squares solver named LSRN that is based on random normal projection, which proves that the preconditioned system is well-conditioned, with a strong concentration result on the extreme singular values, and that the number of iterations is fully predictable when LSQR or the Chebyshev semi-iterative method is applied. Expand
An augmented LSQR method
TLDR
This paper describes how Krylov subspaces generated by the LSQR iterative method can be conveniently augmented with harmonic Ritz vectors. Expand
Convergence of Inner-Iteration GMRES Methods for Rank-Deficient Least Squares Problems
TLDR
A general convergence theory for the generalized minimal residual method preconditioned by inner iterations for solving least squares problems is developed and numerical experiments show that the proposed methods are more robust and efficient compared to previous methods for some rank-deficient problems. Expand
The block LSMR method: a novel efficient algorithm for solving non-symmetric linear systems with multiple right-hand sides
It is well known that if the coefficient matrix in a linear system is large and sparse or sometimes not readily available, then iterative solvers may become the only choice. The block solvers are anExpand
Implicitly restarted Krylov subspace methods for large-scale least-squares problems
The LSQR algorithm is a popular Krylov subspace method for obtaining solutions to large-scale least-squares problems. For some matrices, however, LSQR may require a prohibitively large number ofExpand
A block preconditioner for the Gl-LSMR algorithm
TLDR
A block preconditioner is proposed for the global LSMR method which can be used for solving linear systems with a block partitioned coefficient matrix and multiple right-hand sides and has a better performance in reducing the number of iterations and CPU time. Expand
LSMB: Minimizing the Backward Error for Least-Squares Problems
  • E. Hallman, M. Gu
  • Computer Science, Mathematics
  • SIAM J. Matrix Anal. Appl.
  • 2018
TLDR
An iterative method, LSMB, is given for solving $\min_x \|Ax-b\|_2$ and is constructed so that an objective function closely related to ... Expand
A Hybrid LSMR Algorithm for Large-Scale Tikhonov Regularization
TLDR
A hybrid LSMR algorithm, where Tikhonov regularization is applied to the L SMR subproblem rather than the original problem, is considered, where it is shown that, contrary to standard hybrid methods, hybridLSMR iterates are not equivalent to LSMr iterates on the directly regularized TikhOnov problem. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 32 REFERENCES
LSQR: An Algorithm for Sparse Linear Equations and Sparse Least Squares
TLDR
Numerical tests are described comparing I~QR with several other conjugate-gradient algorithms, indicating that I ~QR is the most reliable algorithm when A is ill-conditioned. Expand
Solving Generalized Least-Squares Problems with LSQR
  • S. Benbow
  • Mathematics, Computer Science
  • SIAM J. Matrix Anal. Appl.
  • 1999
TLDR
Numerical experiments comparing LSQR(A-1)with similar preconditioned Krylov methods are described which demonstrate that the new method exhibits superior numerical properties when the Schur complement BTA-1B is ill conditioned. Expand
Estimating the Backward Error in LSQR
TLDR
This work reviews why the 2-norm of the projection of the residual vector onto the range of $A$ is a useful measure of convergence, and shows how this projection can be estimated efficiently at every iteration of LSQR. Expand
MINRES-QLP: A Krylov Subspace Method for Indefinite or Singular Symmetric Systems
TLDR
This work derives preconditioned MINRES-QLP, new stopping rules, and better estimates of the solution and residual norms, the matrix norm, and the condition number. Expand
Modified Gram-Schmidt (MGS), Least Squares, and Backward Stability of MGS-GMRES
TLDR
This work shows that MGS-GMRES is backward stable, and uses other new results on MGS and its loss of orthogonality, together with an important but neglected condition number, and a relation between residual norms and certain singular values. Expand
Backward perturbation analysis of least squares problems
This thesis is concerned with backward perturbation analyses of the linear least squares (LS) and related problems. Two theoretical measures are commonly used for assessing the backward errors thatExpand
Least-squares problems, normal equations, and stopping criteria for the conjugate gradient method
The Conjugate Gradient method can be successfully used in solving the symmetric and positive definite normal equations obtained from least-squares problems. Taking into account the results ofExpand
Stopping Criteria for the Iterative Solution of Linear Least Squares Problems
TLDR
It is demonstrated that the stopping criteria commonly used with iterative methods for the solution of the linear least squares problem can in some situations be too conservative, causing any chosen method to perform too many iterations or even fail to detect that an acceptable iterate has been obtained. Expand
Least squares problems
Here we solve linear systems Ax = b that do not have a solution. If b is not in the column space of A, there is no x such that Ax = b. The best we can do is to find a vector y that brings left andExpand
Gmres: a Generalized Minimum Residual Algorithm for Solving
Overton, who showed us how the ideal Arnoldi and GMRES problems relate to more general problems of minimization of singular values of functions of matrices 17]. gmres and arnoldi as matrixExpand
...
1
2
3
4
...