Numerical Analytic Continuation Research Articles

Determination of electronic resonances by analytic continuation using barycentric formula

Numerical analytical continuation of a function is used to determine its complex roots. The analytical continuation is carried out by means of a barycentric formula. From the knowledge of the complex roots the energy and width of shape resonances as well as of quantum virtual states can be determined. The roots are calculated for several realistic models and the results are compared with other approaches. We also explore and discuss a stability of the predicted resonant roots with respect to changes of the perturbation potential.

Spectral functions of the strongly interacting three-dimensional Fermi gas

Computing dynamical properties of strongly interacting quantum many-body systems poses a major challenge to theoretical approaches. Usually, one has to resort to numerical analytic continuation of results on imaginary frequencies, which is a mathematically ill-defined procedure. Here we present an efficient method to compute the spectral functions of the two-component Fermi gas near the strongly interacting unitary limit directly in real frequencies. To this end, we combine the Keldysh path integral that is defined in real time with the self-consistent T-matrix approximation. The latter is known to predict thermodynamic and transport properties in good agreement with experimental observations in ultracold atoms. We validate our method by comparison with thermodynamic quantities obtained from imaginary-time calculations and by transforming our real-time propagators to imaginary time. By comparison with state-of-the-art numerical analytic continuation of the imaginary-time results, we show that our real-time results give qualitative improvements for dynamical quantities. Moreover, we show that no significant pseudogap regime exists in the self-consistent T-matrix approximation above the critical temperature Tc, an issue that has been under significant debate. We close by pointing out the versatile nature of our method as it can be extended to other systems, like the spin- or mass-imbalanced Fermi gas, other Bose-Fermi models, two-dimensional systems, and systems out of equilibrium. Published by the American Physical Society 2024

Real-time diagram technique for instantonic systems

The Schwinger-Keldysh diagram technique is usually involved in the calculation of real-time in-in correlation functions. In the case of a thermal state, one can analytically continue imaginary-time Matsubara correlation functions to real times. Nevertheless, not all real-time correlation functions usually can be obtained by such procedure. Moreover, numerical analytic continuation is an ill-posed problem. Thus, even in the case of a thermal state one may need for the Schwinger-Keldysh formalism. If the potential of a system admits degenerate minima, instantonic effects enter the game, so one should also integrate over the instantonic moduli space, including the one, corresponding to the imaginary time translational invariance. However, the Schwinger-Keldysh closed time contour explicitly breaks such invariance. We argue, that this invariance must be recovered, and show, how it can be done. After that, we construct an extension of the Schwinger-Keldysh diagram technique to instantonic systems and demonstrate it on the example of the first few-point correlation functions.

Exploring the electronic transport, magnetic, and optical properties of strongly correlated systems: A numerical analytical continuation approach

The electronic, magnetic and optical properties of the double perovskites Ca2NiOsO6 and Fe-doped derivative were calculated using the full potential linearized augmented plane wave technique through the GGA + U with PBE exchange correlation functionals. The calculations show that both of the systems are half-metallic with Fe-doped system as a ferromagnet, whereas the undoped system shows the ferrimagnetic ordering. Additionally, the study is extended for calculating the Mott parameters through dynamical mean field theory (DMFT) with the maximum entropy model (MEM). It is found that the MIT model parameters (U, [Formula: see text]) for Ca2NiOsO6 and Ca2Fe[Formula: see text]Ni[Formula: see text]OsO6 systems are (5.7[Formula: see text]eV, 6.0[Formula: see text](eV)[Formula: see text]) and (6.0[Formula: see text]eV, 6.0[Formula: see text](eV)[Formula: see text]), respectively. Furthermore, the calculations agree with optical Drude peak analysis. The optical conductivity, reflectivity, absorptivity, ELOSS function, dielectric function refractive index and sum-rule are also explored in relation to the photoinduced behaviors of the materials.

Determination of the Fermi surface by charge density correlations

The Fermi surface topology plays a crucial role in the study of high-temperature superconductivity cuprates. The conventional method for determining the Fermi surface is the maximum spectral intensity method, which involves numerical analytical continuation of Matsubara data for the one-body Green's function. However, the numerical analytical continuation is sensitive to the noise or the precision of the Matsubara data, and hence is not always reliable. With this in mind, we propose a simple and specific notion as a reference for the Fermi surface. It is the derivative of the momentum distribution function ${n}_{\mathbit{k}}$ with respect to the chemical potential $\ensuremath{\mu}$. Our analysis of the noninteracting system shows that the momentum, at which $d{n}_{\mathbit{k}}/d\ensuremath{\mu}$ takes the maximum value, constitutes a surface, which coincides with the Fermi surface. The relationship between $d{n}_{\mathbit{k}}/d\ensuremath{\mu}$ and the Fermi surface in general cases is also analyzed. In order to numerically demonstrate this relationship, we calculate and compare on the two-dimensional Hubbard model within the HGW method. The results show that, at least in the weak and intermediate coupling regimes, using the notion of $d{n}_{\mathbit{k}}/d\ensuremath{\mu}$ to determine the Fermi surface is reliable. We further studied the cases with model parameters standing for realistic materials, and found the surface determined by $d{n}_{\mathbit{k}}/d\ensuremath{\mu}$ exhibits topological transition with the charge filling, similar to the Fermi surface in the cuprates. We believe that using the derivative of the momentum distribution function with respect to the chemical potential to determine the Fermi surface is a reasonable, efficient, and potentially valuable approach.

Description of a glass transition with immeasurable structural relaxation time

A general problem of studying supercooled liquids and glasses is very long relaxation times that do not allow determining explicitly dynamic characteristics. One of the ways to solve this problem is extrapolating values of some dynamical property (e.g. viscosity) from the temperature range where it can be directly measured (simulated) to the low-temperature region. Such extrapolations are usually contradictory because different fitting functions can give substantially different results. Thus, the development of methods for robust extrapolation is an urgent task especially for molecular dynamic study of glassforming liquids. Here we propose a model-free statistical algorithm for a low-temperature extrapolation of liquid viscosity (diffusion coefficient) and apply it to the problem of determination of glass transition temperature and the temperature Tc where the viscosity formally diverges. Our approach is based on numerical analytical continuation of temperature dependence of the viscosity using Padé approximants and error correction procedures using statistical averaging for the treatment of noisy input data. We tested the method on several glass-forming systems and revealed good stability and predictability. Our extrapolation algorithm is suitable for both numerical and experimental studies of glassformers when it is necessary to descend into the parameter range where structural relaxation times of a liquid are too long to be directly obtained.

Exponential asymptotics of woodpile chain nanoptera using numerical analytic continuation

AbstractTraveling waves in woodpile chains are typically nanoptera, which are composed of a central solitary wave and exponentially small oscillations. These oscillations have been studied using exponential asymptotic methods, which typically require an explicit form for the leading‐order behavior. For many nonlinear systems, such as granular woodpile chains, it is not possible to calculate the leading‐order solution explicitly. We show that accurate asymptotic approximations can be obtained using numerical approximation in place of the exact leading‐order behavior. We calculate the oscillation behavior for Toda woodpile chains, and compare the results to exponential asymptotics based on previous methods from the literature: long‐wave approximation and tanh‐fitting. We then use numerical analytic continuation methods based on Padé approximants and the adaptive Antoulas–Anderson (AAA) method. These methods are shown to produce accurate predictions of the amplitude of the oscillations and the mass ratios for which the oscillations vanish. Exponential asymptotics using an AAA approximation for the leading‐order behavior is then applied to study granular woodpile chains, including chains with Hertzian interactions—this method is able to calculate behavior that could not be accurately approximated in previous studies.

TRIQS/SOM 2.0: Implementation of the stochastic optimization with consistent constraints for analytic continuation

TRIQS/SOM 2.0: Implementation of the stochastic optimization with consistent constraints for analytic continuation

Lattice simulations of the QCD chiral transition at real $\mu_B$

Most lattice studies of hot and dense QCD matter rely on extrapolation from zero or imaginary chemical potentials. The ill-posedness of numerical analytic continuation puts severe limitations on the reliability of such methods. We studied the QCD chiral transition at finite real baryon density with the more direct sign reweighting approach. We simulate up to a baryochemical potential-temperature ratio of \mu_B/T=2.7μB/T=2.7, covering the RHIC Beam Energy Scan range, and penetrating the region where methods based on analytic continuation are unpredictive. This opens up a new window to study QCD matter at finite \mu_BμB from first principles. This conference contribution is based on Ref. [1].

Diagrammatic quantum Monte Carlo toward the calculation of transport properties in disordered semiconductors.

A new diagrammatic quantum Monte Carlo approach is proposed to deal with the imaginary time propagator involving both dynamic disorder (i.e., electron-phonon interactions) and static disorder of local or nonlocal nature in a unified and numerically exact way. The establishment of the whole framework relies on a general reciprocal-space expression and a generalized Wick's theorem for the static disorder. Since the numerical cost is independent of the system size, various physical quantities, such as the thermally averaged coherence, Matsubara one-particle Green's function, and current autocorrelation function, can be efficiently evaluated in the thermodynamic limit (infinite in the system size). The validity and performance of the proposed approach are systematically examined in a broad parameter regime. This approach, combined with proper numerical analytic continuation methods and first-principles calculations, is expected to be a versatile tool toward the calculation of various transport properties, such as mobilities in realistic semiconductors involving multiple electronic energy bands, high-frequency optical and low-frequency acoustic phonons, different forms of dynamic and static disorders, and anisotropy.

Fermi polarons at finite temperature: Spectral function and rf spectroscopy

We present a systematic study of a mobile impurity immersed in a three-dimensional Fermi sea of fermions at finite temperature, by using the standard non-self-consistent many-body $T$-matrix theory that is equivalent to a finite-temperature variational approach with the inclusion of one-particle-hole excitation. The impurity spectral function is determined in the real-frequency domain, avoiding any potential errors due to the numerical analytic continuation in previous $T$-matrix calculations and the small spectral broadening parameter used in variational calculations. In the weak-coupling limit, we find that the quasiparticle decay rate of both attractive and repulsive polarons does not increase significantly with increasing temperature, and therefore Fermi polarons may remain well-defined far above Fermi degeneracy. In contrast, near the unitary limit with strong coupling, the decay rate of Fermi polarons rapidly increase and the quasiparticle picture breaks down close to the Fermi temperature. We analyze in detail the recent ejection and injection radio-frequency (rf) spectroscopy measurements, performed at Massachusetts Institute of Technology (MIT) and at European Laboratory for Non-Linear Spectroscopy (LENS), respectively. We show that the momentum average of the spectral function, which is necessary to account for the observed rf-spectroscopy, has a sizable contribution to the width of the quasiparticle peak in spectroscopy. As a result, the measured decay rate of Fermi polarons could be significantly larger than the calculated quasiparticle decay rate at zero momentum. By take this crucial contribution into account, we find that there is a reasonable agreement between theory and experiment for the lifetime of Fermi polarons in the strong-coupling regime, as long as they remain well-defined.

Lattice simulations of the QCD chiral transition at real baryon density

State-of-the-art lattice QCD studies of hot and dense strongly interacting matter currently rely on extrapolation from zero or imaginary chemical potentials. The ill-posedness of numerical analytic continuation puts severe limitations on the reliability of such methods. Here we use the more direct sign reweighting method to perform lattice QCD simulation of the QCD chiral transition at finite real baryon density on phenomenologically relevant lattices. This method does not require analytic continuation and avoids the overlap problem associated with generic reweighting schemes, so has only statistical but no uncontrolled systematic uncertainties for a fixed lattice setup. This opens up a new window to study hot and dense strongly interacting matter from first principles. We perform simulations up to a baryochemical potential-temperature ratio of $\mu_B/T=2.5$ covering most of the RHIC Beam Energy Scan range in the chemical potential. We also clarify the connection of the approach to the more traditional phase reweighting method.

A Posteriori Fractional Tikhonov Regularization Method for the Problem of Analytic Continuation

In this paper, the numerical analytic continuation problem is addressed and a fractional Tikhonov regularization method is proposed. The fractional Tikhonov regularization not only overcomes the difficulty of analyzing the ill-posedness of the continuation problem but also obtains a more accurate numerical result for the discontinuity of solution. This article mainly discusses the a posteriori parameter selection rules of the fractional Tikhonov regularization method, and an error estimate is given. Furthermore, numerical results show that the proposed method works effectively.

Quantum Monte Carlo study of honeycomb antiferromagnets under a triaxial strain

The honeycomb antiferromagnet under a triaxial strain is studied using the quantum Monte Carlo simulation. The strain dimerizes the exchange couplings near the corners, thus destructs the antiferromagnetic order therein. The antiferromagnetic region is continuously reduced by the strain. For the same strain strength, the exact numerical results give a much smaller antiferromagnetic region than the linear spin-wave theory. We then study the strained $XY$ antiferromagnet, where the magnon pseudo-magnetic field behaves quite differently. The $0$th Landau level appears in the middle of the spectrum, and the quantized energies above (below) it are proportional to $n^{\frac{1}{3}} (n^{\frac{2}{3}})$, which is in great contrast to the equally-spaced ones in the Heisenberg case. Besides, we find the antiferromagnetic order of the $XY$ model is much more robust to the dimerization than the Heisenberg one. The local susceptibility of the Heisenberg case is extracted by the numerical analytical continuation, and no sign of the pseudo-Landau levels is resolved. It is still not sure whether the result is due to the intrinsic problem of the numerical analytical continuation. Thus the existence of the magnon pseudo-Landau levels in the spin-$\frac{1}{2}$ strained Heisenberg Hamiltonian remains an open question. Our results are closely related to the two-dimensional van der Waals quantum antiferromagnets and may be realized experimentally.

Real-Frequency Response Functions at Finite Temperature.

Building on previous developments [A. Taheridehkordi, S. H. Curnoe, and J. P. F. LeBlanc, Phys. Rev. B 99, 035120 (2019); PRBMDO2469-995010.1103/PhysRevB.99.035120A. Taheridehkordi, S. H. Curnoe, and J. P. F. LeBlancPhys. Rev. B101, 125109 (2020); PRBMDO2469-995010.1103/PhysRevB.101.125109A. Taheridehkordi, S. H. Curnoe, and J. P. F. LeBlancPhys. Rev. B102, 045115 (2020)PRBMDO2469-995010.1103/PhysRevB.102.045115, B. Holm and U. von Barth, Phys. Rev. B 57, 2108 (1998)PRBMDO0163-182910.1103/PhysRevB.57.2108, J. Vičičević and M. Ferrero, Phys. Rev. B 101, 075113 (2020)PRBMDO2469-995010.1103/PhysRevB.101.075113], we show that the diagrammatic MonteCarlo technique allows us to compute finite-temperature response functions directly on the real-frequency axis within any field-theoretical formulation of the interacting fermion problem. There are no limitations on the type and nature of the system's action or whether partial summation and self-consistent treatment of certain diagram classes are used. In particular, by eliminating the need for numerical analytic continuation from a Matsubara representation, our scheme allows us to study spectral densities of arbitrary complexity with controlled accuracy in models with frequency-dependent effective interactions. For illustrative purposes we consider the problem of the plasmon linewidth in a homogeneous electron gas (jellium).

Deep reinforcement learning for complex evaluation of one-loop diagrams in quantum field theory.

In this paper we present a technique based on deep reinforcement learning that allows for numerical analytic continuation of integrals that are often encountered in one-loop diagrams in quantum field theory. To extract certain quantities of two-point functions, such as spectral densities, mass poles or multiparticle thresholds, it is necessary to perform an analytic continuation of the correlator in question. At one-loop level in Euclidean space, this results in the necessity to deform the integration contour of the loop integral in the complex plane of the square of the loop momentum, to avoid nonanalyticities in the integration plane. Using a toy model for which an exact solution is known, we train a reinforcement learning agent to perform the required contour deformations. Our study shows great promise for an agent to be deployed in iterative numerical approaches used to compute nonperturbative two-point functions, such as the quark propagator Dyson-Schwinger equation, or more generally, Fredholm equations of the second kind, in the complex domain.

Pairing Correlations across the Superfluid Phase Transition in the Unitary Fermi Gas.

In the two-component Fermi gas with a contact interaction, a pseudogap regime can exist at temperatures between the superfluid critical temperature T_{c} and a temperature T^{*}>T_{c}. This regime is characterized by pairing correlations without superfluidity. However, in the unitary limit of infinite scattering length, the existence of this regime is still debated. To help address this, we have applied finite-temperature auxiliary-field quantum MonteCarlo (AFMC) methods to study the thermodynamics of the superfluid phase transition and signatures of the pseudogap in the spin-balanced homogeneous unitary Fermi gas. We present results at finite filling factor ν≃0.06 for the condensate fraction, an energy-staggering pairing gap, the spin susceptibility, and the heat capacity, and compare them to experimental data when available. In contrast to previous AFMC simulations, our model space consists of the complete first Brillouin zone of the lattice, and our calculations are performed in the canonical ensemble of fixed particle number. The canonical ensemble AFMC framework enables the calculation of a model-independent gap, providing direct information on pairing correlations without the need for numerical analytic continuation. We use finite-size scaling to estimate T_{c} at the corresponding filling factor. We find that the energy-staggering pairing gap vanishes above T_{c}, showing no pseudogap effects, and that the spin susceptibility shows a substantially reduced signature of a spin gap compared to previously reported AFMC simulations.

Smooth Self-energy in the Exact-diagonalization-based Dynamical Mean-field Theory: Intermediate-representation Filtering Approach

We propose a method for estimating smooth real-frequency self-energy in the dynamical mean-field theory with the finite-temperature exact diagonalization (DMFT-ED). One of the benefits of DMFT-ED calculations is that one can obtain real-frequency spectra without a numerical analytic continuation. However, these spectra are spiky and strongly depend on the way to discretize a continuous bath (e.g., the number of the bath sites). The present scheme is based on a recently proposed compact representation of imaginary-time Green's functions, the intermediate representation (IR). The projection onto the IR basis acts as a noise filter for the discretization errors in the self-energy. This enables to extract the physically relevant part from the noisy self-energy. We demonstrate the method for single-site DMFT calculations of the single-band Hubbard model. We also show the results can be further improved by numerical analytic continuation.

Numerical analytic continuation of Euclidean data

In this work we present a direct comparison of three different numerical analytic continuation methods: the Maximum Entropy Method, the Backus–Gilbert method and the Schlessinger point or Resonances Via Padé method. First, we perform a benchmark test based on a model spectral function and study the regime of applicability of these methods depending on the number of input points and their statistical error. We then apply these methods to more realistic examples, namely to numerical data on Euclidean propagators obtained from a Functional Renormalization Group calculation, to data from a lattice Quantum Chromodynamics simulation and to data obtained from a tight-binding model for graphene in order to extract the electrical conductivity.

Handling Analysis and Defining Conditions of Dangerous-Sfe Divergent Stability Loss of a Two-Link Road Train Nonlinear Model

Road train steady motion mode divergent stability loss compiles with the critical according to A.M. Liapunov case of a single zero root. That said both safe and dangerous stability loss scenarios are possible according to N.N. Bautin. Dangerous stability loss is followed with a semi-trailer intensive drifting even in case of linear motion. Analyzing the reasons of such system behavior requires developing new effective analytical approaches towards defining safe-dangerous articulated vehicle divergent stability loss because direct methods for finding corresponding Liapunov indexes may appear ineffective in the analytical form being excessively cumbersome. The work presents a formalized approach to analyzing safe stability loss conditions the essence of which is in defining conditions when bifurcation set structure rearrangement occurs in linear motion critical speed small neighborhood. The kind of approach has been tested by the authors when analyzing single unit vehicle stability. Analytical relations are presented defining road train configuration following circular paths with constant Ackermann angle; consideration of analytical results accuracy evaluation is performed based on comparing to the results received with numerical analytic parameter continuation method; analytical relations are received corresponding to safe linear motion mode stability loss (in the sense of N.N. Bautin). The work develops methods of analyzing two-link vehicle non-linear model two-parameter steady modes manifold stability.