betaFeedback

WorksPhysics and Astronomy

A universal black-box quantum Monte Carlo approach to quantum phase transitions

Abstract

We derive exact, universal, closed-form quantum Monte Carlo estimators for finite-temperature energy susceptibility and fidelity susceptibility, applicable to essentially arbitrary Hamiltonians. Combined with recent advancements in Monte Carlo, our approach enables a black-box framework for studying quantum phase transitions–without requiring prior knowledge of an order parameter or the manual design of model-specific ergodic quantum Monte Carlo update rules. We demonstrate the utility of our method by applying a single implementation to the transverse-field Ising model, the XXZ model, and an ensemble of models related by random unitaries.

Introduction

Formally, a quantum phase transition (QPT) is defined by nonanalytic behavior of a system’s ground-state energy in the thermodynamic limit [1]. Traditionally, QPTs are analyzed using local order parameters, such the average local magnetization in the transverse-field Ising model (TFIM) [1]. However, identifying a suitable local order parameter for a new model is often challenging—and for topological QPTs, it is entirely infeasible [2, 3]. Consequently, substantial effort has been devoted to developing generic methods for studying QPTs that do not rely on prior knowledge of an order parameter [4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 20, 21, 22] [17, 18, 19]. Nevertheless, these approaches still rely on manually encoding subtle, system-specific prior knowledge. Building on recent advancements in quantum Monte Carlo (QMC) methods [23, 24, 25] [26] and the well-established fidelity-based approach to QPTs [4, 22] [17, 18, 19], we propose an automated, order-parameter-free, black-box method for studying QPTs that requires no manual encoding of system-specific details.

More precisely, we derive QMC estimators for energy susceptibility (ES) [27, 28] [18] and fidelity susceptibility (FS) [4, 9] [17, 18, 19, 29], which can identify quantum critical points in QPTs without relying on order parameters. While related estimators have appeared in other QMC frameworks [17–19], our derivations and resulting estimators are developed within the relatively recent permutation matrix representation QMC (PMR-QMC), which offers several key advantages over earlier methods [23, 24, 25] [26, 30, 31, 32]. Most relevant to this work, it was recently shown that ergodic PMR-QMC update rules satisfying detailed balance can be generated via a deterministic, automated algorithm for essentially arbitrary Hamiltonians [23, 24, 25]. Furthermore, our estimators are applicable to arbitrarily complex driving terms, as we demonstrate in this work. To the best of our knowledge, neither of these capabilities was supported by previous approaches. Taken together, these features allow our method to be applied directly to a wide range of systems without the need to manually encode laborious, system-specific details.

To this end, we apply our approach to estimate ES and FS in three spin systems: a 2D TFIM model, a 2D XXZ model, and a 100-spin model with random Pauli operator terms that—to our knowledge—cannot be studied using any other method. Given any spin-1/2 Hamiltonian expressed as a sum of Pauli operator products, our current implementation [33] can attempt to locate quantum critical points by estimating ES and/or FS. As with any QMC method, our approach may encounter convergence issues due to frustration or the sign problem [34, 35], which can sometimes be mitigated with custom QMC updates [36, 37] or other techniques [34, 35, 38, 39, 40, 41, 42]. For the three models studied in this work, however, no manual, system-specific adjustments were required to ensure convergence. In all cases, our code automatically monitors the severity of the sign problem [23, 43] and performs thermalization and autocorrelation diagnostics to ensure reliable estimates.

Our implementation [33] builds upon preexisting open-source code for arbitrary spin-1/2 models [43]. Since our estimators are derived within the general PMR-QMC framework, which can represent arbitrary Hamiltonians [26, 30], they can readily be ported to any existing or future PMR-QMC codebases. At present, this includes Bose-Hubbard models on arbitrary graphs [24], arbitrary spin-1/2, high-, and mixed-spin systems [23, 25], and fermionic systems [25, 44]. In summary, PMR-QMC can automatically generate QMC moves for essentially any Hamiltonian [23, 24, 25], and in this work, we derive generic ES and FS estimators within this framework. Combined with automatic convergence diagnostics, our approach—unlike previous numerical methods—can reliably be used as a black-box tool for studying QPTs.

Results

Before deriving formal estimators in Section IV, we demonstrate our method through a series of numerical experiments using open source PMR-QMC code [33] that builds upon prior work [23, 43] (see Section A for basics of PMR-QMC and Section B for basic code instructions). Each model is a spin-1/2 system defined in terms of the standard Pauli operators XX, YY, and ZZ.

We use the notation for the ES χE\chi_E and the FS χF\chi_F as defined in “Theoretical background: Energy and fidelity susceptibilities” in Section IV; the PMR-QMC conventions follow “PMR-QMC background and notation” also in Section IV. XiX_i, YiY_i, and ZiZ_i denote Pauli XX, YY, and ZZ operators acting on spin ii, respectively. All error bars and continuous error bands estimate the spread of the plotted quantity over independent QMC runs: Either 2σ2\sigma (for thermal averages) or direct 95% percentile intervals (for wall-clock times).

We begin with illustrative experiments on well-known models, emphasizing agreement with expected results and, where applicable, reproducing prior plots [18, 19]. We then present a more challenging experiment involving an ensemble of random models, for which our approach still accurately estimates ES and FS. Notably, for this latter case, we are unaware of any other method capable of studying the model successfully. Yet, our single code can handle all of these models with only straightforward adjustments to input files [33].

A simple two spin model

We begin by studying,

H(λ)=Z1Z2+γ(X1+X2)+λ(Z1+Z2),(1)H(\lambda) = Z_1 Z_2 + \gamma(X_1 + X_2) + \lambda(Z_1 + Z_2), \tag*{(1)}

first introduced and studied in a quantum sensing experiment on an NMR system via the fidelity approach (we choose γ=0.1\gamma= 0.1) [45]. By design, this model exhibits an avoided level crossing near the critical points λ≈±1\lambda\approx\pm1, where the ES and FS are expected to peak. Our QMC approach correctly locates these peaks by computing the FS as shown in Figure 1. In fact, as shown, our

Figure 1. FS for the two spin model, Eq. (1), considered in Ref. [45] for different inverse temperatures, β\beta. Data from QMC are averaged over 5 independent random seeds, whereas direct numerical calculations are shown as lines.

approach correctly matches direct numerical calculations for each temperature tested. Furthermore, our plot shows expected convergence of χFβ→χF\chi_F^\beta\to\chi_F as β→∞\beta\to\infty. At β=20\beta= 20 specifically, the QMC results closely match the exactly computed T=0T = 0 results. This verifies the basic veracity of our method.

Next, we remark that prior QMC approaches are generally incapable of estimating the FS, as they must typically restrict themselves to models for which H1∝diag⁡(H)H_1 \propto\operatorname{diag}(H) or H1∝H−diag⁡(H)H_1 \propto H - \operatorname{diag}(H) [22] [17, 18, 19]. Despite the simplicity of this model, it does not satisfy this condition. Yet, our method easily handles this model without even needing to write a specialized code for this case.

A square lattice TFIM

Secondly, we consider the well-known and extensively studied [18, 46, 47] transverse-field Ising model (TFIM),

HTFIM(h)=HJ+hHh=−∑⟨i,j⟩XiXj−h∑i=1nZi,(2)H_{\mathrm{TFIM}}(h)=H_J+hH_h=-\sum_{\langle i,j\rangle}X_iX_j-h\sum_{i=1}^{n}Z_i, \tag*{(2)}

on a square lattice with periodic boundary conditions (PBC). For our purposes, we focus our efforts on a simple 4×44 \times4 model at fixed h=1h=1 and compute the ES and FS as a function of inverse temperature for either HJH_J or hHhhH_h as the “driving term.” Notably, these results are actually not unconditional ES and FS as defined in Eqs. (20) and (21), but rather, the result of restricting computations only to the positive parity subspace of the TFIM. We refer interested readers to Section C for details on this methodological trick, but we remark that our approach is a non-trivial convenience of PMR-QMC that highlights its versatility.

The results are shown in Figure 2 and demonstrate three things. Firstly, they showcase the ability of our method to consider different driving term choices as well as compute things in a fixed parity subspace. Secondly, they help verify the correctness of our approach, as the FS plot correctly matches results independently calculated with SSE-QMC [18]. Finally, they correctly reveal an expected feature of the ES and FS. Namely, one might naively expect that, for the purposes of detecting a quantum critical point, one is free to view either the static term HJH_J or the driving term hHhhH_h in Eq. (2) as the perturbation.

Figure 2. We compute ES and FS restricted to the positive parity subspace (see Section C) of a 4 ×\times 4 square TFIM with PBC and h=1h = 1 as a function of β\beta. QMC data are averaged over 100 independent runs. Our FS results are consistent with a similar computation performed with SSE-QMC [18].

In the T=0T=0 limit, this is correct since

χF,0HJ=h2χF,0hHh,χE,0HJ=h2χE,0hHh,(3)\chi_{F,0}^{H_J}=h^2\chi_{F,0}^{hH_h}, \qquad\chi_{E,0}^{H_J}=h^2\chi_{E,0}^{hH_h}, \tag*{(3)}

follows from plugging in HJ=H−hHhH_J=H-hH_h into Eqs. (9) and (14) alongside the observation ⟨ψn∣H∣ψ0⟩=0\langle\psi_n|H|\psi_0\rangle=0 for all n≠0n\ne0. In these plots, we have chosen h=1h=1, and correspondingly, we find that both the ES and FS converge for either perturbation choice (HJH_J or hHhhH_h) when β\beta is large enough. Yet for any finite β\beta, the relation is less trivial, and in fact, one finds that there is no simple relation that holds across all inverse temperatures [18]. Methodologically, this provides a different way to test for empirical convergence of the finite temperature ES and FS to groundstate values.

So far, we have considered simple models of fixed size. To connect these ideas to the study of QPTs in the infinite system limit, we remark that one can simply compute the ES and FS for increasingly larger system sizes. This is done, for example, in Fig. 4 of Ref. [18]. As expected, this results in an increasingly sharper peak in both the ES and FS as hh approaches the known critical point hc≈3.044h_c\approx3.044 [47]. In addition, one can perform finite size scaling and data collapse in accordance with expected scaling relations, Eqs. (16) and (17), to extract the critical exponent ν≈0.6301\nu\approx0.6301 [18, 46]. Expressed differently, our method is straightforwardly compatible with studying QPTs in the infinite size limit with enough computational effort, which we show more directly for the XXZ model next.

A square lattice XXZ model

Thirdly, we study the antiferromagnetic XXZ model on a square lattice with PBC,

H=Jz4∑⟨i,j⟩ZiZj+λ4∑⟨i,j⟩(XiXj+YiYj),(4)H=\frac{J_z}{4}\sum_{\langle i,j\rangle}Z_iZ_j+\frac{\lambda}{4}\sum_{\langle i,j\rangle}\left(X_iX_j+Y_iY_j\right), \tag*{(4)}

which relates to a hard-core boson model [48, 49] via the Matsubara–Matsuda transformation [50]. The finite-temperature phase diagram of this model has been studied intensely [48], and various facts of this diagram have been replicated by FS approaches with exact diagonalization [51] and SSE-QMC [19]. From a QPT perspective, this model prefers Néel order in the XY plane when λ\lambda dominates and an antiferromagnetic Ising groundstate when JzJ_z dominates. The point λ=Jz\lambda=J_z is a known quantum critical point known as the Heisenberg point [19, 48]. Furthermore [19, 48], it is known that thermal fluctuations at finite temperature destroy the antiferromagnetic Ising phase at a second-order phase transitions. Fixing λ=1\lambda=1, Jz=1.5J_z=1.5 and varying temperature TT, one can search for the expected critical point (T/λ)c≈0.75(T/\lambda)_c\approx0.75 [19, 48]. Using the ES and FS, we search for this critical point in Figure 3, and the empirical peak as we scale system size agrees with expectations. The FS plot in particular is consistent with results obtained in Ref. [19]. More broadly, these experiments show that our method can scale to large system sizes (which are necessary for finite size scaling in the study of QPTs) and also can study finite temperature phase transitions.

Figure 3. We compute ES and FS for the XXZ model in Eq. (4) on a square lattice with PBC as a function of temperature, TT, for different model sizes. QMC data are averaged over at least 200 independent runs. Our FS results are consistent with a similar computation performed with SSE-QMC [19].

A random ensemble

Finally, we consider an ensemble of random 100-spin Hamiltonians generated by random unitary rotations of Eq. (1) (see Section D for full details of this family). By unitary invariance, observables for these random models agree with the simple two qubit model despite each HUi=UiH(λ)Ui†H_{U_i}=U_iH(\lambda)U_i^\dagger and associated driving term UiH1Ui†U_iH_1U_i^\dagger being quite complicated, containing hundreds of random Pauli terms over 100 spins. Using standard QMC approaches, it is not possible to study this ensemble, but it is straightforward to try in PMR-QMC, and the results agree with known expectations as shown in Figure 4.

Figure 4. Comparison of QMC estimates with exact values for an ensemble of 100-spin Hamiltonians generated by random rotations of Eq. (1). Detailed information on the ensemble can be found in Section D. QMC estimates are averaged across 10 random models. In all cases, the QMC estimates show excellent agreement with exact finite β\beta curves. (a) The imaginary time correlator, G(τ)G(\tau), has the expected symmetry around β/2=10\beta/2=10. (b) The ES correctly predicts critical points despite the β=20\beta=20 curve not quite converging to the β=∞\beta=\infty curve. (c) The FS correctly predicts critical points more sharply than ES, and the β=20\beta=20 curve matches the β=∞\beta=\infty curve.

Specifically, PMR-QMC estimators consistently show excellent agreement with exact numerical results for the imaginary-time correlator G(τ)G(\tau), the ES, and the FS, as shown in Figure 4. Furthermore, our results confirm established characteristics of these observables; for instance, G(τ)G(\tau) exhibits symmetry about β/2\beta/2 (see Fig. 1 in Ref. [18] and Fig. 7 in Ref. [19]).

Simulation resources and scaling

Numerical demonstrations for Figure 1 and 4, corresponding to studies of Eq. (1) and its random rotations, were performed on a single 2.3 GHz Intel i9 laptop processor. The remaining experiments involving the TFIM and XXZ models (Figure 2 and 3) were executed on a University of Southern California high-performance computing

cluster. The algorithm is highly memory-efficient [23, 52], with modest requirements even for large-scale simulations (approximately 200 MB of memory per run was sufficient for all calculations reported here). Consequently, our performance analysis focuses primarily on time.

As a QMC method, our approach is naturally suited for lazy parallel execution and exhibits near-perfect strong scaling as the number of cores increases (see, e.g., Fig. 8 of Ref. [23]). Table 1 summarizes the number of MPI threads (NpN_p, approximately corresponding to the number of cores) and wall-clock times used to produce the results in Figure 1 to 4. Additional details regarding simulation parameters and computational resources are provided in Section B.

Fig.NpN_pmin⁡(Twall)\min(T_{\mathrm{wall}})max⁡(Twall)\max(T_{\mathrm{wall}})
1510s1m
210015s14.4hrs
3200–4003.6hrs8.8days
454s8.2m

Table 1. We summarize the resources used to generate data in this work. Further information is contained in Section B.

Next, we briefly discuss the empirical scaling of our algorithm. In Figure 5, we plot the total simulation time used to generate Figure 1 in our study of the two qubit model in Eq. (1). Though a simple model, simulation time grows as a function of β\beta and near the avoided level crossings, as expected.

Figure 5. Simulation time used to generate Figure 1.

In Figure 6, we repeat our study of the XXZ model at the critical temperature Tc≈0.75T_{\mathrm{c}} \approx0.75 at (a) different system sizes N=L2N = L^{2} and (b) with different measurements protocols. We also show the modest wall-clock time scaling as a function of TT for different LL. In our timing experiments, all simulations are run on the same 2.6 GHz AMD EPYC 7513 CPU and simulation parameters are fixed to values listed in the caption, chosen to be in the same range as those used for Figure 3. The different measurements correspond to either running QMC without performing any measurement, while measuring only the ES, or while measuring only the FS. In all cases, times are similar and the scaling is roughly proportional to N2=L4N^{2} = L^{4}. Given that ES and FS have O(1)O(1) and O(q3)O(q^{3}) estimators (see Theorems 1 and 2), this may be surprising, but measurements are performed sparsely (every Nskip=5000N_{\mathrm{skip}} = 5000 MC moves), so wall-clock time is dominated by QMC simulation itself. Further discussion and empirical testing of resource scaling is given in Section B.

Figure 6. Estimate of total simulation time scaling for the L×LL \times L XXZ model data in Figure 3. Timings are given with the following simulation parameters fixed: Ntherm=0N_{\mathrm{therm}} = 0, Nmeas=4×108N_{\mathrm{meas}} = 4 \times10^{8}, Nskip=5×103N_{\mathrm{skip}} = 5 \times10^{3}. Errors bars are estimated over 20 independent runs (Np=20N_{\mathrm{p}} = 20). We refer readers to Section B for more details on the meaning of simulation parameters. (Left) Wall-clock time as a function of LL at the critical temperature Tc≈0.75T_{\mathrm{c}} \approx0.75, separated by what is being measured. Simulation time scales roughly as L4=N2L^{4} = N^{2} in this case. (Right) Wall-clock times to simulate the XXZ model as a function of TT for different LL, with all measurement approaches averaging together.

Discussion

We have developed a universal black-box, finite-temperature quantum Monte Carlo (QMC) approach for studying quantum phase transitions (QPTs). Specifically, we derived PMR-QMC estimators for energy susceptibility (ES) and fidelity susceptibility (FS), which serve as general-purpose, order-parameter-free tools for locating quantum critical points. At the same time, given

a Hamiltonian, we can compute a set of ergodic QMC updates that satisfy detailed balance.

Thanks to the generality of both the PMR-QMC framework and our PMR-based derivations, we successfully studied various aspects of QPTs, ES, and FS for a transverse-field Ising model, an XXZ model, and even a random ensemble—all using a single code. These demonstrations required no model-specific modifications to the source code—only a specification of the Hamiltonian and the desired partition, both of which can be arbitrary. As such, our method and publicly available open-source code are designed to be accessible and user friendly.

Of note, our method can estimate the ES in constant time for models with a simple driving term such as in the TFIM or XXZ model. Though the ES is expected to be a weaker indicator of criticality than the FS in general, we have found it to be sufficient in all models studied in this work. Hence, our approach is not only general but also very efficient for many systems of interest in condensed matter physics.

There are several promising avenues for future research. Foremost, we aim to apply our approach to physically relevant models with non-standard driving terms. We also intend to extend our method to bosonic, fermionic, and high-spin systems of current interest.

We hope the results of this work will provide a valuable and versatile tool for condensed matter physicists and researchers studying quantum phase transitions.

Methods

We introduce requisite technical background then state and discuss the novel PMR-QMC estimators utilized in our study of QPTs in § 2. Importantly, we state Theorems 1 and 2, which provide estimators for ES and FS applicable to most well-studied condensed matter models. At the cost of some efficiency, we also generalize these theorems to arbitrary Hamiltonians.

Energy and fidelity susceptibilities

We discuss the ES and the FS as two generic, order-parameter-free tools to study QPTs [4, 9] [18, 19]. Throughout, we assume an nn spin Hamiltonian with non-degenerate groundstate,

H(λ)=H0+λH1,(5)H(\lambda)=H_0+\lambda H_1, \tag*{(5)}

that has quantum critical point (QCP) at λ=λc\lambda=\lambda_c, denoting the boundary between two quantum phases of a QPT. Henceforth, we refer to H0H_0 as the static term and H1H_1 as the driving term for convenience. Given the spectral decomposition, H(λ)∣ψk(λ)⟩=Ek(λ)∣ψk(λ)⟩H(\lambda)\lvert\psi_k(\lambda)\rangle=E_k(\lambda)\lvert\psi_k(\lambda)\rangle, we define the ES [27, 28] [18]

χE(λ)=−∂2E0(λ)∂λ2.(6)\chi_E(\lambda)=-\frac{\partial^2 E_0(\lambda)}{\partial\lambda^2}. \tag*{(6)}

For second-order QPTs such as that in the TFIM [1], the ES is expected to have a peak or even diverge, and hence, it can be used to locate QCPs. Similarly, the second-order term in the Taylor expansion of groundstate fidelity, or FS [9],

χF(λ)=lim⁡δλ→0−2ln⁡∣⟨ψ0(λ)∣ψ0(λ+δλ)⟩∣δλ2,(7)\chi_F(\lambda)=\lim_{\delta\lambda\to0}\frac{-2\ln\left|\langle\psi_0(\lambda)|\psi_0(\lambda+\delta\lambda)\rangle\right|}{\delta\lambda^2}, \tag*{(7)}

also exhibits a peak or diverges at QCPs for a wide variety of QPTs [28, 53] [18], including first-order [11], second-order [11], topological [54, 55], and as seen recently, even dynamical QPTs [11].

There are several equivalent characterizations of both quantities that are useful in different contexts. First, we discuss various rich characterizations of the FS [4, 7, 9, 28, 53]. By a direct Taylor expansion and simplification of ⟨ψ0(λ)∣ψ0(λ+δλ)⟩\langle\psi_0(\lambda)|\psi_0(\lambda+\delta\lambda)\rangle, one finds the differential form,

References

  1. [1]Sachdev, S. (2011). Quantum Phase Transitions: 2nd ed..
  2. [2]X.-G. Wen, Colloquium: Zoo of quantum-topological phases of matter, Rev. Mod. Phys. 89, 041004 (2017).DOI
  3. [3]A. Hamma, W. Zhang, S. Haas, and D. A. Lidar, Entanglement, fidelity, and topological entropy in a quantum phase transition to topological order, Phys. Rev. B 77, 155111 (2008).DOI
  4. [4]S.-J. Gu, Fidelity approach to quantum phase transitions, Int. J. Mod. Phys. B 24, 4371 (2010).DOI
  5. [5]P. Zanardi and N. Paunković, Ground state overlap and quantum phase transitions, Phys. Rev. E 74, 031123 (2006).DOI
  6. [6]P. Zanardi, L. Campos Venuti, and P. Giorda, Bures metric over thermal state manifolds and quantum criticality, Phys. Rev. A 76, 062318 (2007).DOI
  7. [7]P. Zanardi, P. Giorda, and M. Cozzini, Information-Theoretic Differential Geometry of Quantum Phase Transitions, Phys. Rev. Lett. 99, 100603 (2007).pubmed.ncbi.nlm.nih.gov/17930382
  8. [8]P. Zanardi, H. T. Quan, X. Wang, and C. P. Sun, Mixed-state fidelity and quantum criticality at finite temperature, Phys. Rev. A 75, 032109 (2007).DOI
  9. [9]W.-L. You, Y.-W. Li, and S.-J. Gu, Fidelity, dynamic structure factor, and susceptibility in critical phenomena, Phys. Rev. E 76, 022101 (2007).DOI
  10. [10]M. Rigol, B. S. Shastry, and S. Haas, Fidelity and superconductivity in two-dimensional t - J models, Phys. Rev. B 80, 094529 (2009).DOI
  11. [11]Kazantkin, E., Mozgunov, E., Ezzell, N., & Lidar, D. (2024). Detecting quantum and classical phase transitions via unsupervised machine learning of the fisher information metric. arXiv:2408.03418
  12. [12]G. Sun, A. K. Kolezhuk, and T. Vekua, Fidelity at Berezinskii-Kosterlitz-Thouless quantum phase transitions, Phys. Rev. B 91, 014418 (2015).DOI
  13. [13]H.-Q. Zhou, R. Orús, and G. Vidal, Ground State Fidelity from Tensor Network Representations, Phys. Rev. Lett. 100, 080601 (2008).pubmed.ncbi.nlm.nih.gov/18352611
  14. [14]J. J. Jiang, Y. J. Liu, F. Tang, and C. H. Yang, Reduced fidelity, entanglement and quantum phase transition in the one-dimensional bond-alternating S = 1 Heisenberg chain, Eur. Phys. J. B 83, 1 (2011).DOI
  15. [15]J. Jordan, R. Orús, and G. Vidal, Numerical study of the hard-core Bose-Hubbard model on an infinite square lattice, Phys. Rev. B 79, 174515 (2009).DOI
  16. [16]B. Wang, M. Feng, and Z.-Q. Chen, Berezinskii-Kosterlitz-Thouless transition uncovered by the fidelity susceptibility in the XXZ model, Phys. Rev. A 81, 064301 (2010).DOI
  17. [20]B. Damski, Fidelity susceptibility of the quantum Ising model in the transverse field: The exact solution, Phys. Rev. E 87, 052131 (2013).DOI
  18. [21]M. Białonczyk, F. J. Gómez-Ruiz, and A. del Campo, Uhlmann Fidelity and Fidelity Susceptibility for Integrable Spin Chains at Finite Temperature: Exact Results, New J. Phys. 23, 093033 (2021).DOI
  19. [22]J. Sirker, Finite-Temperature Fidelity Susceptibility for One-Dimensional Quantum Systems, Phys. Rev. Lett. 105, 117203 (2010).DOI
  20. [23]L. Barash, A. Babakhani, and I. Hen, Quantum Monte Carlo algorithm for arbitrary spin-1/2 Hamiltonians, Phys. Rev. B 106, 013281 (2024).DOI
  21. [24]E. Akaturk and I. Hen, Quantum Monte Carlo algorithm for Bose-Hubbard models on arbitrary graphs, Phys. Rev. B 109, 134519 (2024).DOI
  22. [25]A. Babakhani, L. Barash, and I. Hen, A quantum Monte Carlo algorithm for arbitrary high-spin Hamiltonians, preprint arXiv:2503.08039 (2025).DOI
  23. [27]S. Chen, L. Wang, Y. Hao, and Y. Wang, Intrinsic relation between ground-state fidelity and the characterization of a quantum phase transition, Phys. Rev. A 77, 032111 (2008).DOI
  24. [28]L. Campos Venuti and P. Zanardi, Quantum Critical Scaling of the Geometric Tensors, Phys. Rev. Lett. 99, 095701 (2007).DOI
  25. [34]I. Hen, Determining quantum Monte Carlo simulability with geometric phases, Phys. Rev. Res. 3, 023080 (2021).DOI
  26. [35]G. Pan and Z. Y. Meng, The sign problem in quantum Monte Carlo simulations, in Encyclopedia of Condensed Matter Physics (Second Edition), Vol. 1, edited by T. Chakraborty (Academic Press, Oxford, 2024) pp. 879–893.DOI
  27. [36]D. Kandel, R. Ben-Av, and E. Domany, Cluster dynamics for fully frustrated systems, Phys. Rev. Lett. 65, 941 (1990).pubmed.ncbi.nlm.nih.gov/10043065
  28. [37]G. Rakala and K. Damle, Cluster algorithms for frustrated two-dimensional ising antiferromagnets via dual worm constructions, Phys. Rev. E 96, 023304 (2017).DOI
  29. [38]L. Gupta and I. Hen, Elucidating the Interplay between Non-Stoquasticity and the Sign Problem, Adv. Quantum Technol. 3, 1900108 (2020).arxiv.org/abs/1910.13867
  30. [39]I. Hen, Resolution of the sign problem for a frustrated triplet of spins, Physical Review E 99, 033306 (2019).DOI
  31. [40]R. G. Melko, Stochastic series expansion quantum Monte Carlo, in Strongly Correlated Systems: Numerical Methods, Springer Ser. Solid-State Sci., Vol. 176, edited by A. Avella and F. Mancini (Springer, Berlin, Heidelberg, 2013) pp. 185–206.DOI
  32. [41]A. W. Sandvik, Stochastic Series Expansion methods, in Many-Body Methods for Real Materials Modeling and Simulation, Vol. 9, edited by E. Pavarini, E. Koch, and S. Zhang (Verlag des Forschungszentrum Jülich, 2019) pp. 16.1–16.32, arXiv:1909.10591.arxiv.org/abs/1909.10591
  33. [42]Z.-X. Li and H. Yao, Sign-Problem-Free Fermionic Quantum Monte Carlo: Developments and Applications, Annual Review of Condensed Matter Physics 10, 337 (2019).DOI
  34. [43]Permutation matrix representation quantum Monte Carlo for arbitrary spin-1/2 Hamiltonians: program code in c++. (2024). Permutation matrix representation quantum Monte Carlo for arbitrary spin-1/2 Hamiltonians: program code in c++. <a href="https://doi.org/https://github.com/LevBarash/PMRQMC" rel="noopener noreferrer" target="_blank">https://doi.org/https://github.com/LevBarash/PMRQMC</a>
  35. [44]Babakhani, A., Barash, L., & Hen, I. (2025). A quantum monte carlo algorithm for fermionic systems (in preparation).
  36. [45]J. Zhang, X. Peng, N. Rajendran, and D. Suter, Detection of quantum critical points by a probe qubit, Phys. Rev. Lett. 100, 100501 (2008).DOI
  37. [52]L. Gupta, L. Barash, and I. Hen, Calculating the divided differences of the exponential function by addition and removal of inputs, Comput. Phys. Commun. 254, 107385 (2020).DOI
  38. [53]S.-J. Gu, H.-M. Kwok, W.-Q. Ning, and H.-Q. Lin, Fidelity susceptibility, scaling, and universality in quantum critical phenomena, Phys. Rev. B 77, 245109 (2008).DOI
  39. [54]D. F. Abasto, A. Hamma, and P. Zanardi, Fidelity analysis of topological quantum phase transitions, Phys. Rev. A 78, 010301 (2008).DOI
  40. [55]M.-F. Yang, Ground-state fidelity in one-dimensional gapless models, Phys. Rev. B 76, 180403 (2007).DOI
  41. [56]Continentino, M. (2017). Quantum scaling in many-body systems.
  42. [57]C. de Boor, Divided Differences, Surv. Approximation Theory 1, 46 (2005).arxiv.org/abs/math/0502036
  43. [58]A. McCurdy, K. C. Ng, and B. N. Parlett, Accurate computation of divided differences of the exponential function, Math. Comput. 43, 501 (1984).DOI
  44. [59]Q. Zeng, N. Ezzell, A. Babakhani, I. Hen, and L. Barash, Inequalities, identities, and bounds for divided differences of the exponential function, arXiv preprint arXiv:2510.10724 (2025).arXiv
  45. [60]Y.-H. Chen, A. Kalev, and I. Hen, Quantum algorithm for time-dependent hamiltonian simulation by permutation expansion, PRX Quantum 2, 030342 (2021).DOI
  46. [61]A. Kalev and I. Hen, An integral-free representation of the dyson series using divided differences, New Journal of Physics 23, 103035 (2021).DOI
  47. [62]Kalev, A., & Hen, I. (2021). Quantum algorithm for simulating hamiltonian dynamics with an off-diagonal series expansion. Quantum, 5, 426.
  48. [63]A. Kalev and I. Hen, Feynman path integrals for discrete-variable systems: Walks on hamiltonian graphs, Phys. Rev. Res. 7, 013220 (2025).DOI
  49. [64]L. Barash, S. Güttel, and I. Hen, Calculating elements of matrix functions using divided differences, Comput. Phys. Commun. 271, 108219 (2022).DOI
  50. [66]B. A. Berg, Introduction to markov chain monte carlo simulations and their statistical analysis, arXiv preprint cond-mat/0410490 (2004).arXiv
  51. [67]A. F. Izmaylov, T.-C. Yen, R. A. Lang, and V. Verteletskyi, Unitary partitioning approach to the measurement problem in the variational quantum eigensolver method, J. Chem. Theory Comput. 16, 190 (2020).DOI
  52. [68]R. Sarkar and E. v. d. Berg, On sets of commuting and anticommuting paulis, preprint arXiv:1909.08123 (2019).arXiv

Paper details

vlnc:W_5cf5x8bf
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 8

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents