English

Efficient implementations of the modified Gram-Schmidt orthogonalization with a non-standard inner product

Numerical Analysis 2017-03-31 v1

Abstract

The modified Gram-Schmidt (MGS) orthogonalization is one of the most well-used algorithms for computing the thin QR factorization. MGS can be straightforwardly extended to a non-standard inner product with respect to a symmetric positive definite matrix AA. For the thin QR factorization of an m×nm \times n matrix with the non-standard inner product, a naive implementation of MGS requires 2n2n matrix-vector multiplications (MV) with respect to AA. In this paper, we propose nn-MV implementations: a high accuracy (HA) type and a high performance (HP) type, of MGS. We also provide error bounds of the HA-type implementation. Numerical experiments and analysis indicate that the proposed implementations have competitive advantages over the naive implementation in terms of both computational cost and accuracy.

Keywords

Cite

@article{arxiv.1703.10440,
  title  = {Efficient implementations of the modified Gram-Schmidt orthogonalization with a non-standard inner product},
  author = {Akira Imakura and Yusaku Yamamoto},
  journal= {arXiv preprint arXiv:1703.10440},
  year   = {2017}
}

Comments

19 pages, 8 figures