English

Matrix inversion polynomials for the quantum singular value transformation

Quantum Physics 2025-07-22 v1

Abstract

Quantum matrix inversion with the quantum singular value transformation (QSVT) requires a polynomial approximation to 1/x1/x. Several methods from the literature construct polynomials that achieve the known degree complexity O(κlog(κ/ε))\mathcal{O}(\kappa\log(\kappa/\varepsilon)) with condition number κ\kappa and uniform error ε\varepsilon. However, the \emph{optimal} polynomial with lowest degree for fixed error ε\varepsilon can only be approximated numerically with the resource-intensive Remez method, leading to impractical preprocessing runtimes. Here, we derive an analytic shortcut to the optimal polynomial. Comparisons with other polynomials from the literature, based on Taylor expansion, Chebyshev iteration, and convex optimization, confirm that our result is optimal. Furthermore, for large κlog(κ/ε)\kappa\log(\kappa/\varepsilon), our polynomial has the smallest maximum value on [1,1][-1,1] of all approaches considered, leading to reduced circuit depth due to the normalization condition of QSVT. With the Python code provided, this paper will also be useful for practitioners in the field.

Cite

@article{arxiv.2507.15537,
  title  = {Matrix inversion polynomials for the quantum singular value transformation},
  author = {Christoph Sünderhauf and Zalán Németh and Adnaan Walayat and Andrew Patterson and Bjorn K. Berntson},
  journal= {arXiv preprint arXiv:2507.15537},
  year   = {2025}
}
R2 v1 2026-07-01T04:11:10.638Z