Exponentially Decaying Quantum Simulation Error with Noisy Devices

Jue Xu, Chu Zhao, Junyu Fan, Qi Zhao · 2025

Read Exponentially Decaying Quantum Simulation Error with Noisy Devices by Jue Xu, Chu Zhao, Junyu Fan, Qi Zhao (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-10247.txt.

Introduction

Jue Xu and co-authors proved in 2025 a hopeful fact about simulating quantum systems on today's noisy machines. Deepen the circuit, and both the hardware error and the stepping error shrink exponentially per added depth. That runs against the usual dread that depth only accumulates noise. Their analysis covers the standard Trotter approach to simulation.

Simulations across several systems and noise types confirmed the decay. The team then derives the best step count and the noise ceiling that together guarantee a target precision. A phase diagram maps safe territory in depth and noise rate. For fault-tolerant designs, the improved bounds promise real resource savings. The authors frame the work as a systematic guide to practical quantum advantage through simulation.

Catalog record

CREATORJue Xu, Chu Zhao, Junyu Fan, Qi Zhao
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE86 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-10247

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Exponentially Decaying Quantum Simulation Error with Noisy Devices Jue Xu, Chu Zhao, Junyu Fan, Qi Zhao Abstract Quantum simulation is a promising way toward practical quantum advantage, but noise in current quantum hardware poses a significant obstacle. We prove that not only the physical error but also the algorithmic error in a single Trotter step decreases exponentially with the circuit depth. This theoretical finding is validated by our numerical results over various Hamiltonians, initial states, and noise channels. Furthermore, we derive the optimal number of Trotter steps and the noise requirement to guarantee total simulation precision. To explicitly show the requirements for robust quantum simulation, we plot a phase diagram of the accumulated error in terms of circuit depth and noise rate. At last, we demonstrate that our improved error analysis leads to significant resource-saving for fault-tolerant Trotter circuits. By addressing these aspects, this work provides fresh and systematic insight on the practical quantum advantage through quantum simulation. I Introduction Quantum dynamics simulation is a cornerstone application of quantum computation [ 1 , 2 , 3 ] and is widely recognized as a leading candidate for achieving practical quantum advantage [ 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 ] . Efficient quantum dynamics simulation would render general studies of quantum physics tractable [ 12 , 13 , 14 ] , which are notoriously difficult for classical computers. Furthermore, quantum dynamics simulation also serves as an indispensable subroutine for a plethora of quantum algorithms, such as probing ground state energies [ 15 ] , solving linear algebra problems [ 16 , 17 ] , and optimization problems [ 18 , 19 ] . Since the seminal digital quantum dynamics simulation algorithm based on the Trotter formula for local Hamiltonians proposed by Lloyd [ 12 ] , novel post-Trotter techniques [ 20 , 21 , 22 , 23 , 24 ] have been developed and are asymptotically optimal in key parameters. Nonetheless, the variants of the Trotter formula [ 25 , 26 , 27 , 28 , 29 , 30 ] remain primary schemes for near-term quantum devices to achieve practical quantum advantage, owing to their mild hardware requirements and decent theoretical performance. Despite the extensive analyses of Trotter error for showing quantum advantage [ 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 ] , there still lacks systematic analysis of noisy Trotter circuits. Although quantum dynamics simulation is as powerful as universal quantum computation that implies quantum advantage [ 1 ] , its full potential necessitates fault-tolerant quantum computing. Notably, ubiquitous noise detrimentally affects quantum computations, making them less accurate as systems scale up and increasingly susceptible to classical simulation. While a fully fault-tolerant quantum computer remains on the horizon, many quantum simulation experiments have been recently implemented [ 40 , 6 , 11 ] on noisy intermediate-scale quantum (NISQ) devices [ 41 , 42 ] . Significantly, Refs. [ 9 ] simulated the dynamics of the Ising model by Trotterized circuits, but this advantage was challenged by classical simulation techniques, including tensor-network methods for shallow circuits [ 43 , 44 ] and the Pauli path integral method [ 45 , 46 , 47 ] as shown in Fig. 1 . Therefore, it is imperative to determine the noise requirement for an accurate Trotter simulation with potential quantum advantage. Even in the fault-tolerant quantum computing era, with the help of quantum error correction techniques, it is also essential to know the required logical noise rate in a Trotter simulation task. Figure 1: The regime of potential quantum advantage by quantum simulation. For low-depth quantum circuits or strong noise circuits, the quantum advantage would be diminished by classical algorithms. The accuracy of a noisy Trotter simulation task is affected by two competing factors, i.e., circuit depth and noise rate. Specifically, increasing the Trotter steps (circuit depth) suppresses the Trotter error (algorithmic error) but introduces more physical error induced by circuit noise. Although Refs. [ 48 , 49 , 50 ] have examined both physical and algorithmic errors in noisy Trotter circuits, their worst-case error analyses quantified by operator distance between ideal evolution and Trotterized unitaries yield pessimistic estimation due to overlooking crucial evolved state information. Essentially, circuit noise that drives evolved states towards the maximally mixed state causes empirical errors deviating from the worst-case error analyses [ 51 , 52 , 53 ] , resulting in an overestimation of the resources required for an accurate Trotter simulation task. Consequently, to pave the way for demonstrating quantum advantage through Trotter simulation, a refined and state-dependent error analysis is indispensable yet remains unexplored. Our work demonstrates the robustness of Trotter simulation with noisy circuits through several novel contributions. First, we prove that both algorithmic and physical errors in noisy Trotter simulation decay exponentially with Trotter steps (circuit depth) when evolved state information is appropriately considered. We validate this unexpected noise-error relationship through various numerical experiments. As practical applications of our findings, we determine the optimal Trotter number of noisy circuits and establish the noise requirements to guarantee simulation precision, providing quantitative insights into the trade-off between algorithmic and physical errors. As a result, in medium-scale fault-tolerant Trotter simulations, our analysis can reduce the resource requirement (i.e., the circuit depth times the number of physical qubits) by up to 60%, compared with the worst-case analyses. Finally, we present a phase diagram of the total error in a noisy Trotter circuit, delineating the regime of potential quantum advantage. II Results II.1 Setting: noisy Trotter circuit We embark on studying the robustness of noisy Trotter simulation with the following setting. Given an n -qubit local Hamiltonian H=\sum_{l=1}^{L}H_{l} with L=\textup{poly}(n) Pauli terms acting on at most k qubits and evolution time t , quantum dynamics simulation aims to approximate the real-time evolution operator U(t):=e^{-\textup{i}Ht} by a quantum circuit. For this goal, the first-order Lie-Trotter formula \tilde{U}_{1} is a product of the dynamics of each term, that is \addcontentsline{lla}{section}{\numberline q:pf1}\tilde{U}_{1}(\delta t):=e^{-\textup{i}H_{1}\delta t}e^{-\textup{i}H_{2}\delta t}\cdots\,e^{-\textup{i}H_{L}\delta t}=\prod^{L}_{l=1}e^{-\textup{i}H_{l}\delta t}, (1) where each e^{-\textup{i}H_{l}\delta t} is assumed to be implemented efficiently by elementary quantum gates and \delta t=t/r is a short time sliced by r segments [ 12 ] . This type of formula is also called product formula (denoted as PF) in a more general sense. As the Hamiltonian terms are not commutative to each other in general, the algorithmic error introduced by the Trotter formula is called Trotter error. And it has the upper bound \norm{\pf_1(\deltat) - U(\deltat)}_{\mathrm{\infty}}=\mathcal{O}(\delta t^{2}) measured by the spectral norm \norm{\cdot}_{\mathrm{\infty}} . Consequently, the accumulated error between ideal evolution U(t) and r -step Trotter circuit \tilde{U}_{1}^{r}(t/r) scales as \mathcal{O}(t^{2}/r) , which can be suppressed by increasing r . Besides, the Trotter error can also be improved by patterns of products. For instance, the second-order Suzuki-Trotter formula (PF2), i.e., \tilde{U}_{2}(\delta t):=\tilde{U}_{1}(\delta t/2)\tilde{U}_{1}^{\leftarrow}(\delta t/2) , has the error scales \mathcal{O}(\delta t^{3}) , where \tilde{U}_{1}^{\leftarrow} denotes the product in the reverse order. More generally, the p th-order product formulas (PF p ) are defined recursively as \addcontentsline{lla}{section}{\numberline q:high_{p}f}\tilde{U}_{p}(\delta t):=\tilde{U}_{p-2}(u_{p}\delta t)^{2}\tilde{U}_{p-2}((1-4u_{p})\delta t)\tilde{U}_{p-2}(u_{p}\delta t)^{2}, (2) where u_{p}:=1/(4-4^{1/(p-1)}) and p is a positive even integer [ 54 ] . In this way, the Trotter steps of the

Read the complete text