Reduced and Mixed-Precision Block Householder QR Factorization

Peter Benner; Martin Köhler; Xiaobo Liu; Eda Oktay

Home/Archive/arXiv

arXiv · 2610.38537

Reduced and Mixed-Precision Block Householder QR Factorization

Peter Benner·Martin Köhler·Xiaobo Liu·Eda Oktay

View original source ↗Read source document ↗Export metadata ↓

Abstract

Recent advances in hardware architecture have highlighted the importance of reduced-precision arithmetic for improving performance and reducing energy consumption in large-scale scientific applications. However, uniformly low-precision methods can suffer from numerical instability, especially in fundamental linear algebra routines. Mixed-precision strategies therefore play an important role in balancing efficiency, stability, and accuracy. In this paper, we focus on the block Householder QR decomposition using the WY representation and study its behavior in reduced- and mixed-precision settings. Starting from a uniform half-precision variant, we show that performing norm computations with fp32 accumulation is necessary to avoid overflow and underflow on large matrices. Based on this observation, we introduce mpWYQR, a mixed-precision algorithm that increases the precision of only selected computations and avoids most of the precision conversions required by the existing mixed-precision blocked QR algorithm (mpBQR3) introduced in [L. M. Yang, A. Fox, and G. Sanders, SIAM J. Sci. Comp., 43.3 (2021), pp. A1723--A1753]. Our analysis shows that mpWYQR accuracy is governed by low-precision unit roundoff, and that its error bound exceeds that of mpBQR3 by only a constant factor, with identical heuristic bounds. In experiments across a range of matrix sizes and condition numbers, we show that mpWYQR retains mpBQR3 accuracy while running up to $18\times$ faster and using less memory. We also discuss the performance cost of computing inner products in higher precision, and the library support needed for an efficient GPU implementation.

Explore related subjects

math.NA cs.NA

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Peter Benner, Martin Köhler, Xiaobo Liu, Eda Oktay. 2026-10-28. Reduced and Mixed-Precision Block Householder QR Factorization. https://arxiv.org/abs/2610.38537

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

A continuous scale space of diffeomorphisms

In this paper, we define and study a nested family of reproducing kernel Hilbert spaces of vector fields that is indexed by a range of scales, from which we construct a reproducing kernel Hilbert space of scale-dependent vector fields. We provide a characterization of the reproducing kernel of that space, with numerical approximations ensuring quick evaluations when this kernel does not have a closed form. We then introduce a multiscale version of the large deformation diffeomorphic metric mapping (LDDMM) problem and prove the existence of solutions. Finally, we provide numerical experiments performing landmark matching using multiscale LDDMM.

math.NA↗

arXiv2026

Hierarchical Matrices with Tucker Blocks: Construction and Forward Evaluation

In this paper, we introduce hierarchical matrices with Tucker blocks (HMTB) for data-sparse approximation of matrices arising from nonoscillatory and mildly oscillatory kernels. Building on the hierarchical matrix framework, HMTB matrices replace conventional low-rank blocks with Tucker representations. Using high-dimensional interpolation and tensor compression and contractions, we develop algorithms for constructing HMTB matrices and performing matrix-vector multiplication. For an $N \times N$ matrix, these algorithms have linear and quasi-linear complexities in $N$, respectively. Numerical comparisons with an interpolation-based $\mathcal{H}$-matrix implementation demonstrate storage savings and, in several cases, reduced construction times, while maintaining competitive forward-evaluation costs. The framework is also applicable to quasi-uniform grids.

math.NA↗

arXiv2026

Data-driven model order reduction for wave propagation in materials with damage and nonlinearities

In this work, we consider wave propagation in materials characterized by nonlinear properties or damage. To accelerate the simulations of the resulting high-dimensional problems, we apply model order reduction methods. Depending on the knowledge of the underlying equations and the availability of their discrete operators, intrusive methods (here projection-based approaches based on proper orthogonal decomposition (POD)) or non-instrusive methods (here data-driven approaches including dynamic mode decomposition (DMD) and operator inference (OpInf)) can be used. We recall the theoretical foundations of the methods and apply them to the problem of wave propagation. In three different numerical examples, we evaluate the performance of the reduction techniques.

math.NA↗