© 2026 Aqora Quantum S.A.S.

Terms Privacy Legal Notice

arXiv:2410.18736

Error convergence of quantum linear system solvers

M. D. Ginzburg, Ugo Marzolino

Published Oct 24, 2024• 1 citation

Abstract

We analyze the performance of the Harrow-Hassidim-Lloyd algorithm (HHL algorithm) for solving linear problems and of a variant of this algorithm (HHL variant) commonly encountered in literature. This variant relieves the algorithm of preparing an entangled initial state of an auxiliary register. We prove that the computational error of the variant algorithm does not always converge to zero when the number of qubits is increased, unlike the original HHL algorithm. Both algorithms rely upon two fundamental quantum algorithms, the quantum phase estimation and the amplitude amplification. In particular, the error of the HHL variant oscillates due to the presence of undesired phases in the amplitude to be amplified, while these oscillations are suppressed in the original HHL algorithm. Then, we propose a modification of the HHL variant, by amplifying an amplitude of the state vector that does not exhibit the above destructive interference. We also study the complexity of these algorithms in the light of recent results on simulation of unitaries used in the quantum phase estimation step, and show that the modified algorithm has smaller error and lower complexity than the HHL variant. We supported our findings with numerical simulations.

DOI: 10.48550/arxiv.2410.18736

Related papers

Quantum linear systems algorithms: a primer

Danial Dervovic, Mark Herbster, Peter Mountney +3 more

The Harrow-Hassidim-Lloyd (HHL) quantum algorithm for sampling from the solution of a linear system provides an exponential speed-up over its classical counterpart. The problem of solving a system of linear equations has a wide scope of applications, and thus HHL constitutes an important algorithmic primitive. In these notes, we present the HHL algorithm and its improved versions in detail, including explanations of the constituent sub- routines. More specifically, we discuss various quantum subroutines such as quantum phase estimation and amplitude amplification, as well as the important question of loading data into a quantum computer, via quantum RAM. The improvements to the original algorithm exploit variable-time amplitude amplification as well as a method for implementing linear combinations of unitary operations (LCUs) based on a decomposition of the operators using Fourier and Chebyshev series. Finally, we discuss a linear solver based on the quantum singular value estimation (QSVE) subroutine.

arXiv:1802.08227•Feb 22, 2018•62 citations

Error convergence of quantum linear system solvers

M. D. Ginzburg, Ugo Marzolino

The Harrow-Hassidim-Lloyd (HHL) algorithm remains a paradigmatic quantum routine for solving linear systems and a widely used benchmark in theoretical and experimental studies. Several circuit-level variants have been proposed to simplify its implementation, including constructions in which the clock register is factorized via Hadamard gates to reduce circuit depth and entanglement. In this work, we show that this commonly deployed modification can fundamentally alter the asymptotic behavior of the algorithm and may fail to converge in the limit of large clock size due to persistent phase interference effects. This phenomenon is not a finite-size artifact and compromises the correctness guarantees of the factorized-clock implementation. We provide an analytic characterization of the underlying mechanism and demonstrate that a minimal and experimentally feasible modification suffices to restore convergence while preserving the practical advantages of the simplified circuit. Our results clarify an overlooked correctness issue in a widely used HHL implementation pattern and delineate the conditions under which simplified realizations faithfully reproduce the intended algorithmic behavior.

Physical Review A•May 19, 2026•0 citations

On the Phase Estimation Amplitudes in the Quantum Matrix Equation Solver

Xinbo Li, Christopher Phillips, Ian Jeffrey +1 more

The phase estimation amplitude in the quantum matrix equation solver, namely the Harrow/Hassidim/Lloyd (HHL) algorithm is studied in this work. It is the periodic behavior of the amplitude with respect to the eigenvalue estimation error indicates the choice of the simulation time. The impact of the initial states on the quantum phase estimation (QPE) is also provided. This work is expected to be instructive for the formulation of the error analysis in HHL variants.

IEEE Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)•Oct 15, 2023•0 citations

Hybrid quantum linear equation algorithm and its experimental test on IBM Quantum Experience

Yonghae Lee, Jaewoo Joo, Soojoon Lee

We propose a hybrid quantum algorithm based on the Harrow-Hassidim-Lloyd (HHL) algorithm for solving a system of linear equations. In this paper, we show that our hybrid algorithm can reduce a circuit depth from the original HHL algorithm by means of a classical information feed-forward after the quantum phase estimation algorithm, and the results of the hybrid algorithm are identical to those of the HHL algorithm. In addition, it is experimentally examined with four qubits in the IBM Quantum Experience setups, and the experimental results of our algorithm show higher accurate performance on specific systems of linear equations than that of the HHL algorithm.

arXiv:1807.10651•Scientific Reports•Mar 18, 2019•77 citations

Solving the encoding bottleneck: of the HHL algorithm, by the HHL algorithm

Guang Ping He

The Harrow-Hassidim-Lloyd (HHL) algorithm offers exponential speedup for solving the quantum linear-system problem. But some caveats for the speedup could be hard to met. One of the difficulties is the encoding bottleneck, i.e., the efficient preparation of the initial quantum state. To prepare an arbitrary $N$-dimensional state exactly, existing state-preparation approaches generally require a runtime of $O(N)$, which will ruin the speedup of the HHL algorithm. Here we show that the states can be prepared approximately with a runtime of $O(poly(\log N))$ by employing a slightly modified version of the HHL algorithm itself. Thus, applying this approach to prepare the initial state of the original HHL algorithm can preserve the exponential speedup advantage. It can also serve as a standalone solution for other applications demanding fast state preparation.

arXiv:2502.13534•Feb 19, 2025•0 citations

An Enhanced Hybrid HHL Algorithm

Jack Morgan, Éric Ghysels, Hamed Mohammadbagherpoor

We present a classical enhancement to improve the accuracy of the Hybrid variant (Hybrid HHL) of the quantum algorithm for solving linear systems of equations proposed by Harrow, Hassidim, and Lloyd (HHL). We achieve this by using higher precision quantum estimates of the eigenvalues relevant to the linear system, and a new classical step to guide the eigenvalue inversion part of Hybrid HHL. We show that eigenvalue estimates with just two extra bits of precision result in tighter error bounds for our Enhanced Hybrid HHL compared to HHL. Our enhancement reduces the error of Hybrid HHL by an average of 57 percent on an ideal quantum processor for a representative sample of 2x2 systems. On IBM Torino and IonQ Aria-1 hardware, we see that the error of Enhanced Hybrid HHL is on average 13 percent and 20 percent (respectively) less than that of HHL for the same set of systems.

arXiv:2404.10103•Apr 15, 2024•0 citations

We analyze the performance of the Harrow-Hassidim-Lloyd algorithm (HHL algorithm) for solving linear problems and of a variant of this algorithm (HHL variant) commonly encountered in literature. This variant relieves the algorithm of preparing an entangled initial state of an auxiliary register. We prove that the computational error of the variant algorithm does not always converge to zero when the number of qubits is increased, unlike the original HHL algorithm. Both algorithms rely upon two fundamental quantum algorithms, the quantum phase estimation and the amplitude amplification. In particular, the error of the HHL variant oscillates due to the presence of undesired phases in the amplitude to be amplified, while these oscillations are suppressed in the original HHL algorithm. Then, we propose a modification of the HHL variant, by amplifying an amplitude of the state vector that does not exhibit the above destructive interference. We also study the complexity of these algorithms in the light of recent results on simulation of unitaries used in the quantum phase estimation step, and show that the modified algorithm has smaller error and lower complexity than the HHL variant. We supported our findings with numerical simulations.

The Harrow-Hassidim-Lloyd (HHL) quantum algorithm for sampling from the solution of a linear system provides an exponential speed-up over its classical counterpart. The problem of solving a system of linear equations has a wide scope of applications, and thus HHL constitutes an important algorithmic primitive. In these notes, we present the HHL algorithm and its improved versions in detail, including explanations of the constituent sub- routines. More specifically, we discuss various quantum subroutines such as quantum phase estimation and amplitude amplification, as well as the important question of loading data into a quantum computer, via quantum RAM. The improvements to the original algorithm exploit variable-time amplitude amplification as well as a method for implementing linear combinations of unitary operations (LCUs) based on a decomposition of the operators using Fourier and Chebyshev series. Finally, we discuss a linear solver based on the quantum singular value estimation (QSVE) subroutine.

The Harrow-Hassidim-Lloyd (HHL) algorithm remains a paradigmatic quantum routine for solving linear systems and a widely used benchmark in theoretical and experimental studies. Several circuit-level variants have been proposed to simplify its implementation, including constructions in which the clock register is factorized via Hadamard gates to reduce circuit depth and entanglement. In this work, we show that this commonly deployed modification can fundamentally alter the asymptotic behavior of the algorithm and may fail to converge in the limit of large clock size due to persistent phase interference effects. This phenomenon is not a finite-size artifact and compromises the correctness guarantees of the factorized-clock implementation. We provide an analytic characterization of the underlying mechanism and demonstrate that a minimal and experimentally feasible modification suffices to restore convergence while preserving the practical advantages of the simplified circuit. Our results clarify an overlooked correctness issue in a widely used HHL implementation pattern and delineate the conditions under which simplified realizations faithfully reproduce the intended algorithmic behavior.

We propose a hybrid quantum algorithm based on the Harrow-Hassidim-Lloyd (HHL) algorithm for solving a system of linear equations. In this paper, we show that our hybrid algorithm can reduce a circuit depth from the original HHL algorithm by means of a classical information feed-forward after the quantum phase estimation algorithm, and the results of the hybrid algorithm are identical to those of the HHL algorithm. In addition, it is experimentally examined with four qubits in the IBM Quantum Experience setups, and the experimental results of our algorithm show higher accurate performance on specific systems of linear equations than that of the HHL algorithm.

The Harrow-Hassidim-Lloyd (HHL) algorithm offers exponential speedup for solving the quantum linear-system problem. But some caveats for the speedup could be hard to met. One of the difficulties is the encoding bottleneck, i.e., the efficient preparation of the initial quantum state. To prepare an arbitrary

N

-dimensional state exactly, existing state-preparation approaches generally require a runtime of

O(N)

, which will ruin the speedup of the HHL algorithm. Here we show that the states can be prepared approximately with a runtime of

O(poly(\log N))

by employing a slightly modified version of the HHL algorithm itself. Thus, applying this approach to prepare the initial state of the original HHL algorithm can preserve the exponential speedup advantage. It can also serve as a standalone solution for other applications demanding fast state preparation.

We present a classical enhancement to improve the accuracy of the Hybrid variant (Hybrid HHL) of the quantum algorithm for solving linear systems of equations proposed by Harrow, Hassidim, and Lloyd (HHL). We achieve this by using higher precision quantum estimates of the eigenvalues relevant to the linear system, and a new classical step to guide the eigenvalue inversion part of Hybrid HHL. We show that eigenvalue estimates with just two extra bits of precision result in tighter error bounds for our Enhanced Hybrid HHL compared to HHL. Our enhancement reduces the error of Hybrid HHL by an average of 57 percent on an ideal quantum processor for a representative sample of 2x2 systems. On IBM Torino and IonQ Aria-1 hardware, we see that the error of Enhanced Hybrid HHL is on average 13 percent and 20 percent (respectively) less than that of HHL for the same set of systems.