Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2609.35114

Accurate and bounded approximation of quantum correlation functions

Alexander van Lomwel, Florian Mintert

quant-phSubmitted Sep 28, 2026• Announced Sep 29, 2026• 0 citations
arXivHTMLPDF

Abstract

The exact classical simulation of non-integrable quantum systems rapidly becomes intractable beyond small system sizes, yet its role is central to predicting many-body dynamics. Approximate methods alleviate this cost but typically lack rigorous guarantees on their deviation from the exact result. For the infinite-temperature correlation function, we derive a rigorous error bound for approximating the full dynamics within a computable finite-sized region. Applied to autocorrelations, this finite-region approach is especially accurate across one-dimensional systems and beyond, and can further provide rigorous error bounds for other approximate methods, demonstrated here for matrix-product-operator simulations.

AI Summary

The paper addresses how to certify classical approximations to infinite-temperature dynamical correlation functions without simulating the full system. It approximates $C_F(t)=2^{-n}\operatorname{tr}[A_F(t)B]$ by evolving observables inside a finite spatial window and derives an error bound requiring only window dynamics and boundary couplings. For observables supported inside the window and boundary interaction $V$ satisfying $\operatorname{tr}_{\mathrm{out}}V=0$, exact integral identities and Cauchy–Schwarz give $|C_F(t)-C_W(t)|\leq\mathcal B_C(t)$, where $\mathcal B_C(t)=\int_0^t dr\,\|[V(r),A_W(t,r)]\|\int_0^r dp\,\|[V(p),B_W(p)]\|$, using normalized Hilbert–Schmidt norms. If the individual observable errors begin at orders $t^{m_A}$ and $t^{m_B}$, both the correlation error and its bound begin no earlier than $t^{m_A+m_B}$. Autocorrelations therefore have doubled short-time error order. The bound is independent of total system size and applies in the thermodynamic limit. Tests use Ising and Heisenberg chains with windows up to 13 spins, plus honeycomb-lattice Ising windows. Comparison with exactly computable transverse-field Ising dynamics checks bound tightness. For the largest honeycomb window, the correlation bound reaches $10^{-2}$ at $gt=2.011$, versus $0.742$ for the observable-based bound. Other approximations, including MPO simulations, receive the certificate $|C_F-C_A|\leq\mathcal B_C+|C_W-C_A|$. The scope is infinite-temperature correlations. Certification remains limited to times when the finite-window bound is small; larger windows extend this interval, with less spatial reach per spin beyond one dimension. Extensions to OTOCs and full counting statistics are suggested but not established.

Little sign of AI-generated textOverclaimsPossible error

Error found: The MPO figure caption incorrectly claims that |C_A-C_W|<=2B_C guarantees accuracy at least as good as the window approximation, whereas it establishes only that such accuracy is possible. The uniform nearest-neighbor isotropic Heisenberg chain in Eq. (XXX) is integrable, contrary to its repeated description as non-integrable. The mixed-field Ising numerical examples omit the field strengths h_x and h_z needed to reproduce their correlation curves and reported error-bound threshold times.

Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.

Related papers

Real-time dynamics of matrix quantum mechanics beyond the classical approximation

P. V. Buividovich, Masanori Hanada, Andreas Schäfer A

We describe a numerical method which allows to go beyond the classical approximation for the real-time dynamics of many-body systems by approximating the many-body Wigner function by the most general Gaussian function with time-dependent mean and dispersion. On a simple example of a classically chaotic system with two degrees of freedom we demonstrate that this Gaussian state approximation is accurate for significantly smaller field strengths and longer times than the classical one. Applying this approximation to matrix quantum mechanics, we demonstrate that the quantum Lyapunov exponents are in general smaller than their classical counterparts, and even seem to vanish below some temperature. This behavior resembles the finite-temperature phase transition which was found for this system in Monte-Carlo simulations, and ensures that the system does not violate the Maldacena-Shenker-Stanford bound λL < 2πT, which inevitably happens for classical dynamics at sufficiently small temperatures.

•Mar 26, 2018•12 citations

Real-time dynamics of matrix quantum mechanics beyond the classical approximation

P. V. Buividovich, Masanori Hanada, Andreas Schäfer

We describe a numerical method which allows to go beyond the classical approximation for the real-time dynamics of many-body systems by approximating the many-body Wigner function by the most general Gaussian function with time-dependent mean and dispersion. On a simple example of a classically chaotic system with two degrees of freedom we demonstrate that this Gaussian state approximation is accurate for significantly smaller field strengths and longer times than the classical one. Applying this approximation to matrix quantum mechanics, we demonstrate that the quantum Lyapunov exponents are in general smaller than their classical counterparts, and even seem to vanish below some temperature. This behavior resembles the finite-temperature phase transition which was found for this system in Monte-Carlo simulations, and ensures that the system does not violate the Maldacena-Shenker-Stanford bound λL < 2πT, which inevitably happens for classical dynamics at sufficiently small temperatures.

arXiv:1711.05556•EPJ Web of Conferences•Jan 1, 2018•13 citations

Bounding the finite-size error of quantum many-body dynamics simulations

Zhiyuan Wang, Michael Foss‐Feig, Kaden R. A. Hazzard

Finite-size errors (FSEs), the discrepancies between an observable in a finite system and in the thermodynamic limit, are ubiquitous in numerical simulations of quantum many-body systems.Although a rough estimate of these errors can be obtained from a sequence of finite-size results, a strict, quantitative bound on the magnitude of FSE is still missing.Here we derive rigorous upper bounds on the FSE of local observables in real-time quantum dynamics simulations initialized from a product state.In d-dimensional locally interacting systems with a finite local Hilbert space, our bound implies | Ŝ(t ) L -Ŝ(t ) ∞ | C(2vt/L) cL-μ , with v, C, c, μ constants independent of L and t, which we compute explicitly.For periodic boundary conditions (PBCs), the constant c is twice as large as that for open boundary conditions (OBCs), suggesting that PBCs have smaller FSEs than OBCs at early times.The bound can be generalized to a large class of correlated initial states as well.As a byproduct, we prove that the FSE of local observables in ground-state simulations decays exponentially with L under a suitable spectral gap condition.Our bounds are practically useful in determining the validity of finite-size results, as we demonstrate in simulations of the one-dimensional (1D) quantum Ising and Fermi-Hubbard models.

arXiv:2009.12032•Physical Review Research•Aug 16, 2021•9 citations

Dynamical Properties of Quantum Many-Body Systems at Zero Temperature

Eduardo R. Gagliano, C. A. Balseiro

We present a method to compute dynamical correlation functions in quantum many-body systems at zero temperature. The ground state of a finite system is evaluated exactly by a modified Lanczos method and from it the real-frequency correlation functions are obtained by a projective technique for the memory-function formalism. We apply the method to the density-density correlation function of a one-dimensional spinless fermion system. Exact results are recovered in the noninteracting limit; in the strong-coupling limit our results are compared with Monte Carlo simulation and analytical approximations.

Physical Review Letters•Dec 28, 1987•264 citations

Classical Simulability from Operator Entanglement Scaling

Neil Dowling

Local-operator entanglement (LOE) quantifies the nonlocal structure of Heisenberg operators and serves as a diagnostic of many-body chaos. We provide rigorous bounds showing when an operator can be well-approximated by a matrix-product operator (MPO), given asymptotic scaling of its LOE $α$-Rényi entropies. Specifically, we prove that a volume law scaling for $α\geq 1$ implies that the operator cannot be approximated efficiently as an MPO while faithfully reproducing all expectation values. On the other hand, if we restrict to correlations over a relevant sub-class of (ensembles of) states, then logarithmic scaling of the $α< 1$ entropies implies MPO simulability. This result covers a range of relevant quantities, including infinite temperature autocorrelation functions, out-of-time-ordered correlators, and average-case expectation values over ensembles of computational basis states. Beyond this regime, we provide numerical evidence together with a random matrix model to argue that this simulability result also typically holds for arbitrary states. Our results put on firm footing the heuristic expectation that a low operator entanglement implies efficient tensor network representability, extending celebrated foundational results from the theory of matrix-product states and providing a formal link between quantum chaos and classical simulability.

quant-pharXiv:2603.05656•Physical Review A•Mar 9, 2026•15 citations

Recent Advances in the Calculation of Dynamical Correlation Functions

J. Florencio, O. F. de Alcantara Bonfim

We review various theoretical methods that have been used in recent years to calculate dynamical correlation functions of many-body systems. Time-dependent correlation functions and their associated frequency spectral densities are the quantities of interest, for they play a central role in both the theoretical and experimental understanding of dynamic properties. In particular, dynamic correlation functions appear in the fluctuation-dissipation theorem, where the response of a many-body system to an external perturbation is given in terms of the relaxation function of the unperturbed system, provided the disturbance is small. The calculation of the relaxation function is rather difficult in most cases of interest, except for a few examples where exact analytic expressions are allowed. For most of systems of interest approximation schemes must be used. The method of recurrence relation has, at its foundation, the solution of Heisenberg equation of motion of an operator in a many-body interacting system. Insights have been gained from theorems that were discovered with that method. For instance, the absence of pure exponential behavior for the relaxation functions of any Hamiltonian system. The method of recurrence relations was used in quantum systems such as dense electron gas, transverse Ising model, Heisenberg model, XY model, Heisenberg model with Dzyaloshinskii-Moriya interactions, as well as classical harmonic oscillator chains. Effects of disorder were considered in some of those systems. In the cases where analytical solutions were not feasible, approximation schemes were used, but are highly model-dependent. Another important approach is the numericallly exact diagonalizaton method. It is used in finite-sized systems, which sometimes provides very reliable information of the dynamics at the infinite-size limit. In this work, we discuss the most relevant applications of the method of recurrence relations and numerical calculations based on exact diagonalizations. The method of recurrence relations relies on the solution to the coefficients of a continued fraction for the Laplace transformed relaxation function. The calculation of those coefficients becomes very involved and, only a few cases offer exact solution. We shall concentrate our efforts on the cases where extrapolation schemes must be used to obtain solutions for long times (or low frequency) regimes. We also cover numerical work based on the exact diagonalization of finite sized systems. The numerical work provides some thermodynamically exact results and identifies some difficulties intrinsic to the method of recurrence relations.

arXiv:2005.14222•Frontiers in Physics•Nov 26, 2020•12 citations
The exact classical simulation of non-integrable quantum systems rapidly becomes intractable beyond small system sizes, yet its role is central to predicting many-body dynamics. Approximate methods alleviate this cost but typically lack rigorous guarantees on their deviation from the exact result. For the infinite-temperature correlation function, we derive a rigorous error bound for approximating the full dynamics within a computable finite-sized region. Applied to autocorrelations, this finite-region approach is especially accurate across one-dimensional systems and beyond, and can further provide rigorous error bounds for other approximate methods, demonstrated here for matrix-product-operator simulations.
The paper addresses how to certify classical approximations to infinite-temperature dynamical correlation functions without simulating the full system. It approximates CF(t)=2−ntr⁡[AF(t)B]C_F(t)=2^{-n}\operatorname{tr}[A_F(t)B]CF​(t)=2−ntr[AF​(t)B] by evolving observables inside a finite spatial window and derives an error bound requiring only window dynamics and boundary couplings.
For observables supported inside the window and boundary interaction VVV satisfying tr⁡outV=0\operatorname{tr}_{\mathrm{out}}V=0trout​V=0, exact integral identities and Cauchy–Schwarz give ∣CF(t)−CW(t)∣≤BC(t)|C_F(t)-C_W(t)|\leq\mathcal B_C(t)∣CF​(t)−CW​(t)∣≤BC​(t), where BC(t)=∫0tdr ∥[V(r),AW(t,r)]∥∫0rdp ∥[V(p),BW(p)]∥\mathcal B_C(t)=\int_0^t dr\,\|[V(r),A_W(t,r)]\|\int_0^r dp\,\|[V(p),B_W(p)]\|BC​(t)=∫0t​dr∥[V(r),AW​(t,r)]∥∫0r​dp∥[V(p),BW​(p)]∥, using normalized Hilbert–Schmidt norms. If the individual observable errors begin at orders tmAt^{m_A}tmA​ and tmBt^{m_B}tmB​, both the correlation error and its bound begin no earlier than tmA+mBt^{m_A+m_B}tmA​+mB​. Autocorrelations therefore have doubled short-time error order. The bound is independent of total system size and applies in the thermodynamic limit.
Tests use Ising and Heisenberg chains with windows up to 13 spins, plus honeycomb-lattice Ising windows. Comparison with exactly computable transverse-field Ising dynamics checks bound tightness. For the largest honeycomb window, the correlation bound reaches 10−210^{-2}10−2 at gt=2.011gt=2.011gt=2.011, versus 0.7420.7420.742 for the observable-based bound. Other approximations, including MPO simulations, receive the certificate ∣CF−CA∣≤BC+∣CW−CA∣|C_F-C_A|\leq\mathcal B_C+|C_W-C_A|∣CF​−CA​∣≤BC​+∣CW​−CA​∣.
The scope is infinite-temperature correlations. Certification remains limited to times when the finite-window bound is small; larger windows extend this interval, with less spatial reach per spin beyond one dimension. Extensions to OTOCs and full counting statistics are suggested but not established.
We describe a numerical method which allows to go beyond the classical approximation for the real-time dynamics of many-body systems by approximating the many-body Wigner function by the most general Gaussian function with time-dependent mean and dispersion. On a simple example of a classically chaotic system with two degrees of freedom we demonstrate that this Gaussian state approximation is accurate for significantly smaller field strengths and longer times than the classical one. Applying this approximation to matrix quantum mechanics, we demonstrate that the quantum Lyapunov exponents are in general smaller than their classical counterparts, and even seem to vanish below some temperature. This behavior resembles the finite-temperature phase transition which was found for this system in Monte-Carlo simulations, and ensures that the system does not violate the Maldacena-Shenker-Stanford bound λL < 2πT, which inevitably happens for classical dynamics at sufficiently small temperatures.
We describe a numerical method which allows to go beyond the classical approximation for the real-time dynamics of many-body systems by approximating the many-body Wigner function by the most general Gaussian function with time-dependent mean and dispersion. On a simple example of a classically chaotic system with two degrees of freedom we demonstrate that this Gaussian state approximation is accurate for significantly smaller field strengths and longer times than the classical one. Applying this approximation to matrix quantum mechanics, we demonstrate that the quantum Lyapunov exponents are in general smaller than their classical counterparts, and even seem to vanish below some temperature. This behavior resembles the finite-temperature phase transition which was found for this system in Monte-Carlo simulations, and ensures that the system does not violate the Maldacena-Shenker-Stanford bound λL < 2πT, which inevitably happens for classical dynamics at sufficiently small temperatures.
Finite-size errors (FSEs), the discrepancies between an observable in a finite system and in the thermodynamic limit, are ubiquitous in numerical simulations of quantum many-body systems.Although a rough estimate of these errors can be obtained from a sequence of finite-size results, a strict, quantitative bound on the magnitude of FSE is still missing.Here we derive rigorous upper bounds on the FSE of local observables in real-time quantum dynamics simulations initialized from a product state.In d-dimensional locally interacting systems with a finite local Hilbert space, our bound implies | Ŝ(t ) L -Ŝ(t ) ∞ | C(2vt/L) cL-μ , with v, C, c, μ constants independent of L and t, which we compute explicitly.For periodic boundary conditions (PBCs), the constant c is twice as large as that for open boundary conditions (OBCs), suggesting that PBCs have smaller FSEs than OBCs at early times.The bound can be generalized to a large class of correlated initial states as well.As a byproduct, we prove that the FSE of local observables in ground-state simulations decays exponentially with L under a suitable spectral gap condition.Our bounds are practically useful in determining the validity of finite-size results, as we demonstrate in simulations of the one-dimensional (1D) quantum Ising and Fermi-Hubbard models.
We present a method to compute dynamical correlation functions in quantum many-body systems at zero temperature. The ground state of a finite system is evaluated exactly by a modified Lanczos method and from it the real-frequency correlation functions are obtained by a projective technique for the memory-function formalism. We apply the method to the density-density correlation function of a one-dimensional spinless fermion system. Exact results are recovered in the noninteracting limit; in the strong-coupling limit our results are compared with Monte Carlo simulation and analytical approximations.
Local-operator entanglement (LOE) quantifies the nonlocal structure of Heisenberg operators and serves as a diagnostic of many-body chaos. We provide rigorous bounds showing when an operator can be well-approximated by a matrix-product operator (MPO), given asymptotic scaling of its LOE ααα-Rényi entropies. Specifically, we prove that a volume law scaling for α≥1α\geq 1α≥1 implies that the operator cannot be approximated efficiently as an MPO while faithfully reproducing all expectation values. On the other hand, if we restrict to correlations over a relevant sub-class of (ensembles of) states, then logarithmic scaling of the α<1α< 1α<1 entropies implies MPO simulability. This result covers a range of relevant quantities, including infinite temperature autocorrelation functions, out-of-time-ordered correlators, and average-case expectation values over ensembles of computational basis states. Beyond this regime, we provide numerical evidence together with a random matrix model to argue that this simulability result also typically holds for arbitrary states. Our results put on firm footing the heuristic expectation that a low operator entanglement implies efficient tensor network representability, extending celebrated foundational results from the theory of matrix-product states and providing a formal link between quantum chaos and classical simulability.
We review various theoretical methods that have been used in recent years to calculate dynamical correlation functions of many-body systems. Time-dependent correlation functions and their associated frequency spectral densities are the quantities of interest, for they play a central role in both the theoretical and experimental understanding of dynamic properties. In particular, dynamic correlation functions appear in the fluctuation-dissipation theorem, where the response of a many-body system to an external perturbation is given in terms of the relaxation function of the unperturbed system, provided the disturbance is small. The calculation of the relaxation function is rather difficult in most cases of interest, except for a few examples where exact analytic expressions are allowed. For most of systems of interest approximation schemes must be used. The method of recurrence relation has, at its foundation, the solution of Heisenberg equation of motion of an operator in a many-body interacting system. Insights have been gained from theorems that were discovered with that method. For instance, the absence of pure exponential behavior for the relaxation functions of any Hamiltonian system. The method of recurrence relations was used in quantum systems such as dense electron gas, transverse Ising model, Heisenberg model, XY model, Heisenberg model with Dzyaloshinskii-Moriya interactions, as well as classical harmonic oscillator chains. Effects of disorder were considered in some of those systems. In the cases where analytical solutions were not feasible, approximation schemes were used, but are highly model-dependent. Another important approach is the numericallly exact diagonalizaton method. It is used in finite-sized systems, which sometimes provides very reliable information of the dynamics at the infinite-size limit. In this work, we discuss the most relevant applications of the method of recurrence relations and numerical calculations based on exact diagonalizations. The method of recurrence relations relies on the solution to the coefficients of a continued fraction for the Laplace transformed relaxation function. The calculation of those coefficients becomes very involved and, only a few cases offer exact solution. We shall concentrate our efforts on the cases where extrapolation schemes must be used to obtain solutions for long times (or low frequency) regimes. We also cover numerical work based on the exact diagonalization of finite sized systems. The numerical work provides some thermodynamically exact results and identifies some difficulties intrinsic to the method of recurrence relations.