Quantum Krylov Algorithm for Szegö Quadrature

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Kirby, William, Shen, Yizhi, Camps, Daan, Chowdhury, Anirban, Klymko, Katherine, Van Beeumen, Roel
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914052440064000
author Kirby, William
Shen, Yizhi
Camps, Daan
Chowdhury, Anirban
Klymko, Katherine
Van Beeumen, Roel
author_facet Kirby, William
Shen, Yizhi
Camps, Daan
Chowdhury, Anirban
Klymko, Katherine
Van Beeumen, Roel
contents We present a quantum algorithm to evaluate matrix elements of functions of unitary operators. The method is based on calculating quadrature nodes and weights using data collected from a quantum processor. Given a unitary $U$ and quantum states $|ψ_0\rangle$, $|ψ_1\rangle$, the resulting quadrature rules form a functional that can then be used to classically approximate $\langleψ_1|f(U)|ψ_0\rangle$ for any function $f$. In particular, the algorithm calculates Szegö quadrature rules, which, when $f$ is a Laurent polynomial, have the optimal relation between degree of $f$ and number of distinct quantum circuits required. The unitary operator $U$ could approximate a time evolution, opening the door to applications like estimating properties of Hamiltonian spectra and Gibbs states, but more generally could be any operator implementable via a quantum circuit. We expect this algorithm to be useful as a subroutine in other quantum algorithms, much like quantum signal processing or the quantum eigenvalue transformation of unitaries. Key advantages of our algorithm are that it does not require approximating $f$ directly, via a series expansion or in any other way, and once the output functional has been constructed using the quantum algorithm, it can be applied to any $f$ classically after the fact.
format Preprint
id arxiv_https___arxiv_org_abs_2509_19195
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Quantum Krylov Algorithm for Szegö Quadrature
Kirby, William
Shen, Yizhi
Camps, Daan
Chowdhury, Anirban
Klymko, Katherine
Van Beeumen, Roel
Quantum Physics
We present a quantum algorithm to evaluate matrix elements of functions of unitary operators. The method is based on calculating quadrature nodes and weights using data collected from a quantum processor. Given a unitary $U$ and quantum states $|ψ_0\rangle$, $|ψ_1\rangle$, the resulting quadrature rules form a functional that can then be used to classically approximate $\langleψ_1|f(U)|ψ_0\rangle$ for any function $f$. In particular, the algorithm calculates Szegö quadrature rules, which, when $f$ is a Laurent polynomial, have the optimal relation between degree of $f$ and number of distinct quantum circuits required. The unitary operator $U$ could approximate a time evolution, opening the door to applications like estimating properties of Hamiltonian spectra and Gibbs states, but more generally could be any operator implementable via a quantum circuit. We expect this algorithm to be useful as a subroutine in other quantum algorithms, much like quantum signal processing or the quantum eigenvalue transformation of unitaries. Key advantages of our algorithm are that it does not require approximating $f$ directly, via a series expansion or in any other way, and once the output functional has been constructed using the quantum algorithm, it can be applied to any $f$ classically after the fact.
title Quantum Krylov Algorithm for Szegö Quadrature
topic Quantum Physics
url https://arxiv.org/abs/2509.19195