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:2509.15539 · npj Computational Materials

Improving Spectral Resolution from Real-time Evolution for Correlated Systems

Ta Tang, Chunjing Jia, Brian Moritz, Thomas Devereaux

Published Aug 11, 2026• 0 citations
arXivHTMLPDF

Abstract

Abstract The quality of numerically simulated spectra using real-time evolution methods for strongly correlated systems is affected by both the length of simulation time and the system size, limiting resolution in both frequency and momentum. In this work, we propose a computationally cheap, linear autoregressive machine learning-based framework to extend short-time and short-distance results over a wider range. We use the proposed method to extend the lesser Green’s function for both the Hubbard model and the much more computationally challenging Hubbard-extended Holstein model. This technique significantly improves both the frequency and momentum resolution of the single-particle removal spectrum $${\mathcal{A}}(k,\omega )$$ A ( k , ω ) , allowing the observation of otherwise obscured spectral features due to electron-phonon coupling.

DOI
10.1038/s41524-026-02275-8

Related papers

Spectral Decomposition and High-Accuracy Green’s Functions: Overcoming the Nyquist-Shannon Limit via Complex-time Krylov Expansion

Sebastian Paeckel

The accurate computation of low-energy spectra of strongly correlated quantum many-body systems, typically accessed via Green's functions, is a long-standing problem posing enormous challenges to numerical methods. When the spectral decomposition is obtained from Fourier transforming a time series, the Nyquist-Shannon theorem limits the frequency resolution Δω according to the numerically accessible time domain size T via Δω=2π/T. In tensor network methods, increasing the domain size is exponentially hard due to the ubiquitous spread of correlations, limiting the frequency resolution and thereby restricting this ansatz class mostly to one-dimensional systems with small quasiparticle velocities. Here, we show how this limitation can be overcome by augmenting the time series with complex-time Krylov states. With the example of the critical S-1/2 Heisenberg model and light bipolarons in the two-dimensional Su-Schrieffer-Heeger model, we demonstrate the enormous improvements in accuracy, which can be achieved using this method.

arXiv:2411.09680•Physical Review Letters•Mar 16, 2026•2 citations

Machine learning Green’s functions of strongly correlated Hubbard models

Mateo Cárdenes Wuttig

Abstract We demonstrate that a machine learning framework based on kernel ridge regression can encode and predict the self-energy of one-dimensional Hubbard models using only mean-field features such as static and dynamic Hartree–Fock quantities and first-order GW calculations. This approach is applicable across a wide range of on-site Coulomb interaction strengths U / t , ranging from weakly interacting systems ( U / t ≪ 1 ) to strong correlations ( U / t > 8 ). The predicted self-energy is transformed via Dyson’s equation and analytic continuation to obtain the real-frequency Green’s function, which allows access to the spectral function and density of states. This method can be used for nearest-neighbor interactions t and long-range hopping terms t ′, t ′′, and t ′ ′ ′ .

arXiv:2511.07252•Journal of Physics Condensed Matter•Apr 24, 2026•0 citations

A Practical Framework for Simulating Time-Resolved Spectroscopy Based on a Real-Time Dyson Expansion

Cian C. Reeves, Michael Alexander Kurniawan, Yuanran Zhu +5 more

Time-resolved spectroscopy is a powerful tool for probing electron dynamics in molecules and solids, revealing transient phenomena on subfemtosecond time scales. The interpretation of experimental results is often enhanced by parallel numerical studies, which can provide insight and validation for experimental hypotheses. However, developing a theoretical framework for simulating time-resolved spectra remains a significant challenge. The most suitable approach involves the many-body nonequilibrium Green’s function formalism, which accounts for crucial dynamical many-body correlations during time evolution. While these dynamical correlations are essential for observing emergent behavior in time-resolved spectra, they also render the formalism prohibitively expensive for large-scale simulations. Substantial effort has been devoted to reducing this computational cost─through approximations and numerical techniques─while preserving the key dynamical correlations. The ultimate goal is to enable first-principles simulations of time-dependent systems ranging from small molecules to large, periodic, multidimensional solids. In this perspective, we outline key challenges in developing practical simulations for time-resolved spectroscopy, with a particular focus on Green’s function methodologies. We highlight a recent advancement toward a scalable framework: the real-time Dyson expansion (RT-DE) [ Phys. Rev. Lett. 2024, 133, 226902]. We introduce the theoretical foundation of RT-DE and discuss strategies for improving scalability, which have already enabled simulations of system sizes beyond the reach of previous fully dynamical approaches. We conclude with an outlook on future directions for extending RT-DE to first-principles studies of dynamically correlated, nonequilibrium systems.

arXiv:2505.00667•Journal of Chemical Theory and Computation•Jun 27, 2025•4 citations

Maximum-Entropy Extension of Quantum Correlation Functions from Short Real-Time Dynamics

Filippo Maria Gambetta, Sabrina Yue Wang, Raul A. Santos

Correlation functions are central to understanding the behavior of quantum systems, with their spectra giving direct access to quasiparticle and collective excitations. However, the long-time evolution required for sufficient spectral resolution is challenging to reach for both classical and quantum simulations. In this work, we introduce a principled approach for extending time series of quantum correlation functions and reconstructing their spectral densities, based on their connection with covariance sequences of stationary stochastic processes. This connection enables us to apply Burg's maximum-entropy principle to find the least-committal extension compatible with the measured values of the correlation function, which corresponds to an all-pole autoregressive model whose coefficients solve a set of Yule--Walker equations. The resulting extension is positive semi-definite by construction, guaranteeing the positivity of the spectrum and the automatic satisfaction of its total spectral weight sum rule. The approach applies to both scalar and matrix-valued correlation functions, including those with highly structured or continuous spectra. The generalization to the matrix case makes it possible to trade simulation time for additional measured correlation functions, enabling accurate time-series extension and spectral reconstruction from short-time data, even in the presence of noise.

quant-pharXiv:2610.03696•Oct 5, 2026•0 citations

Complex time evolution in tensor networks and time-dependent Green's functions

M. Grundner, Philipp Westhoff, Fabian B. Kugler +2 more

Real-time calculations in tensor networks are strongly limited in time by entanglement growth, restricting the achievable frequency resolution of Green's functions, spectral functions, self-energies, and other related quantities. By extending the time evolution to contours in the complex plane, entanglement growth is curtailed, enabling numerically efficient high-precision calculations of time-dependent correlators and Green's functions with detailed frequency resolution. Various approaches to time evolution in the complex plane and the required postprocessing for extracting the pure real-time and frequency information are compared. We benchmark our results on the examples of the single-impurity Anderson model using matrix product states and of the three-band Hubbard-Kanamori and Dworin-Narath models using a tree tensor network. Our findings indicate that the proposed methods are also applicable to challenging realistic calculations of materials.

arXiv:2312.11705•Physical review. B./Physical review. B•Apr 8, 2024•17 citations

Autoregressive Neural Network Extrapolation of Quantum Spin Dynamics Across Time and Space

Hubert Pugzlys, Shreyas Varude, Sam Dillon +5 more

Understanding the long-time dynamics of strongly correlated quantum matter remains a major challenge: as quantum entanglement grows during time evolution, conventional numerical methods become increasingly costly, limiting access to the long-range correlations and low-energy excitations that govern gapless systems. Here we show that these dynamical correlations can be extended beyond the accessible numerical window without explicitly evolving the underlying many-body wavefunction. We introduce an autoregressive machine-learning framework that enables the extrapolation of dynamical spin correlations in both time and space beyond the reach of conventional numerical methods. Trained on time-dependent density matrix renormalization group simulations of the gapless XXZ model, our approach is benchmarked against exact solutions available for this analytically solvable system. Combined with physics-informed spatial extension, a multi-layer perceptron model using ReLU activation functions has been shown to be superior to convolutional neural networks and linear regressions for longer time extrapolation. A study of error accumulation further demonstrates that our autoregressive neural network extrapolations are highly robust to perturbations, suggesting stable and reliable predictions. This work offers a complementary route to studying the dynamics of gapless quantum many-body systems, in which machine learning extends and complements the capabilities of state-of-the-art numerical approaches.

arXiv:2512.13103•Dec 15, 2025•0 citations
Abstract The quality of numerically simulated spectra using real-time evolution methods for strongly correlated systems is affected by both the length of simulation time and the system size, limiting resolution in both frequency and momentum. In this work, we propose a computationally cheap, linear autoregressive machine learning-based framework to extend short-time and short-distance results over a wider range. We use the proposed method to extend the lesser Green’s function for both the Hubbard model and the much more computationally challenging Hubbard-extended Holstein model. This technique significantly improves both the frequency and momentum resolution of the single-particle removal spectrum A(k,ω){\mathcal{A}}(k,\omega )A(k,ω) A ( k , ω ) , allowing the observation of otherwise obscured spectral features due to electron-phonon coupling.
The accurate computation of low-energy spectra of strongly correlated quantum many-body systems, typically accessed via Green's functions, is a long-standing problem posing enormous challenges to numerical methods. When the spectral decomposition is obtained from Fourier transforming a time series, the Nyquist-Shannon theorem limits the frequency resolution Δω according to the numerically accessible time domain size T via Δω=2π/T. In tensor network methods, increasing the domain size is exponentially hard due to the ubiquitous spread of correlations, limiting the frequency resolution and thereby restricting this ansatz class mostly to one-dimensional systems with small quasiparticle velocities. Here, we show how this limitation can be overcome by augmenting the time series with complex-time Krylov states. With the example of the critical S-1/2 Heisenberg model and light bipolarons in the two-dimensional Su-Schrieffer-Heeger model, we demonstrate the enormous improvements in accuracy, which can be achieved using this method.
Abstract We demonstrate that a machine learning framework based on kernel ridge regression can encode and predict the self-energy of one-dimensional Hubbard models using only mean-field features such as static and dynamic Hartree–Fock quantities and first-order GW calculations. This approach is applicable across a wide range of on-site Coulomb interaction strengths U / t , ranging from weakly interacting systems ( U / t ≪ 1 ) to strong correlations ( U / t > 8 ). The predicted self-energy is transformed via Dyson’s equation and analytic continuation to obtain the real-frequency Green’s function, which allows access to the spectral function and density of states. This method can be used for nearest-neighbor interactions t and long-range hopping terms t ′, t ′′, and t ′ ′ ′ .
Time-resolved spectroscopy is a powerful tool for probing electron dynamics in molecules and solids, revealing transient phenomena on subfemtosecond time scales. The interpretation of experimental results is often enhanced by parallel numerical studies, which can provide insight and validation for experimental hypotheses. However, developing a theoretical framework for simulating time-resolved spectra remains a significant challenge. The most suitable approach involves the many-body nonequilibrium Green’s function formalism, which accounts for crucial dynamical many-body correlations during time evolution. While these dynamical correlations are essential for observing emergent behavior in time-resolved spectra, they also render the formalism prohibitively expensive for large-scale simulations. Substantial effort has been devoted to reducing this computational cost─through approximations and numerical techniques─while preserving the key dynamical correlations. The ultimate goal is to enable first-principles simulations of time-dependent systems ranging from small molecules to large, periodic, multidimensional solids. In this perspective, we outline key challenges in developing practical simulations for time-resolved spectroscopy, with a particular focus on Green’s function methodologies. We highlight a recent advancement toward a scalable framework: the real-time Dyson expansion (RT-DE) [ Phys. Rev. Lett. 2024, 133, 226902]. We introduce the theoretical foundation of RT-DE and discuss strategies for improving scalability, which have already enabled simulations of system sizes beyond the reach of previous fully dynamical approaches. We conclude with an outlook on future directions for extending RT-DE to first-principles studies of dynamically correlated, nonequilibrium systems.
Correlation functions are central to understanding the behavior of quantum systems, with their spectra giving direct access to quasiparticle and collective excitations. However, the long-time evolution required for sufficient spectral resolution is challenging to reach for both classical and quantum simulations. In this work, we introduce a principled approach for extending time series of quantum correlation functions and reconstructing their spectral densities, based on their connection with covariance sequences of stationary stochastic processes. This connection enables us to apply Burg's maximum-entropy principle to find the least-committal extension compatible with the measured values of the correlation function, which corresponds to an all-pole autoregressive model whose coefficients solve a set of Yule--Walker equations. The resulting extension is positive semi-definite by construction, guaranteeing the positivity of the spectrum and the automatic satisfaction of its total spectral weight sum rule. The approach applies to both scalar and matrix-valued correlation functions, including those with highly structured or continuous spectra. The generalization to the matrix case makes it possible to trade simulation time for additional measured correlation functions, enabling accurate time-series extension and spectral reconstruction from short-time data, even in the presence of noise.
Real-time calculations in tensor networks are strongly limited in time by entanglement growth, restricting the achievable frequency resolution of Green's functions, spectral functions, self-energies, and other related quantities. By extending the time evolution to contours in the complex plane, entanglement growth is curtailed, enabling numerically efficient high-precision calculations of time-dependent correlators and Green's functions with detailed frequency resolution. Various approaches to time evolution in the complex plane and the required postprocessing for extracting the pure real-time and frequency information are compared. We benchmark our results on the examples of the single-impurity Anderson model using matrix product states and of the three-band Hubbard-Kanamori and Dworin-Narath models using a tree tensor network. Our findings indicate that the proposed methods are also applicable to challenging realistic calculations of materials.
Understanding the long-time dynamics of strongly correlated quantum matter remains a major challenge: as quantum entanglement grows during time evolution, conventional numerical methods become increasingly costly, limiting access to the long-range correlations and low-energy excitations that govern gapless systems. Here we show that these dynamical correlations can be extended beyond the accessible numerical window without explicitly evolving the underlying many-body wavefunction. We introduce an autoregressive machine-learning framework that enables the extrapolation of dynamical spin correlations in both time and space beyond the reach of conventional numerical methods. Trained on time-dependent density matrix renormalization group simulations of the gapless XXZ model, our approach is benchmarked against exact solutions available for this analytically solvable system. Combined with physics-informed spatial extension, a multi-layer perceptron model using ReLU activation functions has been shown to be superior to convolutional neural networks and linear regressions for longer time extrapolation. A study of error accumulation further demonstrates that our autoregressive neural network extrapolations are highly robust to perturbations, suggesting stable and reliable predictions. This work offers a complementary route to studying the dynamics of gapless quantum many-body systems, in which machine learning extends and complements the capabilities of state-of-the-art numerical approaches.