English

A Backward Stable Algorithm for Computing the CS Decomposition via the Polar Decomposition

Numerical Analysis 2018-04-25 v1

Abstract

We introduce a backward stable algorithm for computing the CS decomposition of a partitioned 2n×n2n \times n matrix with orthonormal columns, or a rank-deficient partial isometry. The algorithm computes two n×nn \times n polar decompositions (which can be carried out in parallel) followed by an eigendecomposition of a judiciously crafted n×nn \times n Hermitian matrix. We prove that the algorithm is backward stable whenever the aforementioned decompositions are computed in a backward stable way. Since the polar decomposition and the symmetric eigendecomposition are highly amenable to parallelization, the algorithm inherits this feature. We illustrate this fact by invoking recently developed algorithms for the polar decomposition and symmetric eigendecomposition that leverage Zolotarev's best rational approximations of the sign function. Numerical examples demonstrate that the resulting algorithm for computing the CS decomposition enjoys excellent numerical stability.

Keywords

Cite

@article{arxiv.1804.09002,
  title  = {A Backward Stable Algorithm for Computing the CS Decomposition via the Polar Decomposition},
  author = {Evan S. Gawlik and Yuji Nakatsukasa and Brian D. Sutton},
  journal= {arXiv preprint arXiv:1804.09002},
  year   = {2018}
}
R2 v1 2026-06-23T01:33:57.332Z