Two-Time Quantum Fluctuations Approach and its Relation to the Bethe--Salpeter Equation

This paper provides a detailed analysis of a two-time quantum fluctuations approach, demonstrating its equivalence to the Bethe-Salpeter equation for two-time exchange-correlation functions when the generalized Kadanoff-Baym ansatz with Hartree-Fock propagators is applied.

The Gist

Imagine you are trying to predict how a massive crowd of people behaves after someone shouts "Fire!" in a stadium. You want to know not just where people are right now, but how the panic ripples through the crowd over time, how groups form, and how they react to each other.

In the world of physics, this "crowd" is a quantum many-particle system (like electrons in a solid or atoms in a cold gas), and the "shout" is an external energy burst. Physicists have been trying to build a perfect "crowd simulator" for decades, but it's incredibly hard because every particle is constantly bumping into and influencing every other particle.

This paper introduces a new, smarter way to run that simulation. Here is the breakdown in simple terms:

1. The Problem: The "Memory" Bottleneck

Traditionally, simulating these systems is like trying to record a movie where every single frame depends on every single previous frame.

• The Old Way (Standard Methods): To know what happens at minute 10, you have to remember exactly what happened at minute 1, 2, 3... all the way up to 9. As the simulation gets longer, the computer has to store a massive amount of data. It's like trying to carry a library in your backpack while running; eventually, you get tired (the computer runs out of memory), and the simulation stops.
• The Cost: This method is accurate but computationally expensive. It scales poorly, meaning if you double the time you want to simulate, the work doesn't just double; it explodes.

2. The New Idea: Watching the "Wiggles"

The authors (Schroedter and Bonitz) propose a different perspective. Instead of tracking every single particle's exact history, they focus on the "fluctuations" or the "wiggles."

Think of a calm lake.

• The Standard Approach tries to track every single water molecule.
• The Fluctuation Approach says, "Let's ignore the calm water and just watch the ripples."

They developed a method called the Quantum Fluctuations Approach. Instead of calculating the entire history of the crowd, they calculate how the deviations (the ripples) from the average behavior evolve.

• The Magic Trick: This new method is "linear." If you want to simulate twice as long, it takes twice as much time, not exponentially more. It's like switching from carrying a library to carrying a single notebook. It's much lighter on the computer's memory.

3. The Big Discovery: Two Roads, One Destination

The paper's main breakthrough is proving that this new "Ripple Method" is actually the same as a very famous, established method called the Bethe-Salpeter Equation (which is part of the GW Approximation).

• The Analogy: Imagine two different maps to get to the same city.
  • Map A (The Old Way): A detailed, winding road map that shows every tree and pothole (The Bethe-Salpeter Equation).
  • Map B (The New Way): A high-speed highway guide that skips the small details but gets you there just as fast (The Quantum Fluctuations Approach).

The authors proved mathematically that if you use a specific set of rules (called the Generalized Kadanoff-Baym Ansatz), these two maps lead to the exact same destination. They showed that the "Ripple Method" isn't just a cheap shortcut; it's a rigorous, accurate way to solve the same complex physics problem.

4. Testing the Theory: The "Stadium" Experiments

To prove their theory works, they ran simulations on a "stadium" made of a grid of spots (called the Hubbard model).

• Small Stadium (6 spots): They compared their new method against the "Gold Standard" (Exact Diagonalization) and a "Stochastic" method (random sampling).
  • Result: For weak interactions (calm crowd), all methods agreed perfectly.
  • Result: For strong interactions (panicked crowd), the new method (and the GW method) stayed accurate longer than the random sampling method, which started to hallucinate wild oscillations.
• Large Stadium (30 spots): As the system got bigger, the new method became even better. It showed that for large systems, the "Ripple Method" and the "Stochastic Method" are practically identical, but the Ripple Method is more stable and easier to understand physically.

5. Why This Matters

This paper is a bridge. It connects a brand-new, efficient computational trick (Quantum Fluctuations) with a classic, heavy-duty theory (Bethe-Salpeter/GW).

• For Scientists: It means we can now simulate complex quantum materials (like superconductors or dense plasmas) for longer times and on larger systems without needing a supercomputer the size of a building.
• The Takeaway: By focusing on the "wiggles" (fluctuations) rather than the whole history, we can predict how quantum systems react to shocks with high accuracy and low cost. It's like realizing you don't need to track every grain of sand in an hourglass to know how much time has passed; you just need to watch the flow.

In a nutshell: The authors found a way to make quantum simulations faster and lighter by focusing on the "ripples" in the system, and they proved that this clever shortcut is mathematically identical to the heavy, traditional way of doing things.

Technical Summary

1. Problem Statement

The simulation of correlated quantum many-particle systems out of equilibrium (e.g., dense plasmas, ultracold atoms, correlated solids) is computationally demanding. While Non-Equilibrium Green Functions (NEGF) provide a rigorous framework, standard simulations suffer from cubic scaling with simulation time ($N_t^3$) and high memory costs due to the storage of two-particle Green functions ($G_2$).

Recent advances, such as the $G_1$–$G_2$ scheme, have achieved linear time scaling ($N_t$) but at the cost of massive memory requirements for storing the full two-particle correlation matrix. Furthermore, while the Quantum Fluctuations Approach (based on stochastic dynamics of fluctuations) was previously shown to be equivalent to the time-local GW approximation, its physical meaning and its relationship to the Bethe–Salpeter Equation (BSE) for two-time (non-local in time) exchange-correlation functions remained unestablished.

2. Methodology

The authors develop a theoretical framework connecting the Two-Time Quantum Fluctuations Approach (specifically the Quantum Polarization Approximation, PA) with the Nonequilibrium GW Approximation derived from the Bethe–Salpeter Equation.

• Theoretical Framework:
  • The study utilizes the Keldysh contour formalism for NEGFs.
  • It decomposes the two-particle NEGF into Hartree-Fock contributions and a correlated part (exchange-correlation function, $L$).
  • It defines single-particle fluctuation operators ($\delta \hat{G}$) and two-particle fluctuation correlations ($L$).
• The Quantum Polarization Approximation (PA):
  • Based on a stochastic hierarchy of fluctuation equations.
  • The core approximation assumes that second-order fluctuations (fluctuations of fluctuations) can be decoupled by replacing them with "source fluctuations" (products of single-particle Green functions).
  • This leads to a closed set of equations of motion (EOMs) for two-time two-particle fluctuations.
• The Bethe–Salpeter Approach (GW):
  • The authors derive EOMs for the two-time correlated Green function ($G$) within the GW approximation using the Generalized Kadanoff-Baym Ansatz (GKBA).
  • Specifically, they employ the Hartree-Fock GKBA (HF-GKBA), which reconstructs off-diagonal time components of the Green function using retarded/advanced propagators and time-diagonal values, avoiding the need to solve full two-time integral equations.
• Derivation of Equivalence:
  • The authors derive the EOMs for the two-time GW approximation and compare them term-by-term with the EOMs of the PA.
  • They demonstrate that under the HF-GKBA and in the weak coupling limit, the PA is mathematically equivalent to the GW approximation (including exchange contributions, denoted as GW$\pm$).

3. Key Contributions

1. Establishment of Equivalence: The paper proves that the Quantum Polarization Approximation is equivalent to the Bethe–Salpeter equation for the two-time exchange-correlation function when the HF-GKBA is applied. This bridges the gap between stochastic fluctuation methods and standard diagrammatic many-body theory.
2. Two-Time Generalization: The authors extend the previously known time-local equivalence (between PA and GW) to the full two-time formulation. This allows for the calculation of dynamic response functions and spectral properties that depend on two distinct time arguments.
3. Physical Interpretation: The work clarifies the physical meaning of the central approximation in the fluctuation approach (the quantum polarization approximation) by mapping it directly to the irreducible vertex in the BSE.
4. Computational Efficiency: The derived equations for the two-time GW approximation retain the favorable linear time scaling ($N_t$) of the $G_1$–$G_2$ scheme but are formulated in a way that is compatible with stochastic methods. This avoids the $N_b^6$ scaling (where $N_b$ is basis size) and massive memory costs associated with storing full rank-four tensors in traditional GW implementations.

4. Numerical Results

The authors validated the theory using the Fermi-Hubbard model on 1D chains with periodic boundary conditions, following a confinement quench (sudden expansion).

• System Sizes: Simulations were performed on chains with $N=6$ and $N=30$ sites.
• Comparisons: Results from the Two-Time PA and Two-Time GW were compared against:
  • Exact Diagonalization (for $N=6$).
  • Stochastic Polarization Approximation with Multiple Ensembles (SPA-ME).
• Findings:
  • Weak Coupling ($U=0.1J$): Excellent agreement was found between PA, GW, and exact solutions for both small and large systems.
  • Stronger Coupling ($U=0.5J$):
    • For the small system ($N=6$), PA and GW showed good frequency agreement but overestimated oscillation amplitudes compared to exact diagonalization. The GW approximation showed slightly better damping behavior than PA.
    • For the large system ($N=30$), the PA and SPA-ME were found to be fully equivalent, confirming that the multiple-ensemble stochastic extension is negligible for larger systems.
    • The GW approximation remained more stable than PA at strong coupling, exhibiting smaller amplitude deviations, though both approximations struggled to capture the strong damping seen in exact solutions (a known limitation of HF-GKBA).
  • Time Dependence: The agreement between approximations and exact solutions degrades as the time difference ($t - t'$) increases, particularly for stronger interactions, highlighting the sensitivity of long-time propagation to the approximation scheme.

5. Significance

• Theoretical Unification: The paper provides a rigorous theoretical link between stochastic fluctuation dynamics and the standard diagrammatic GW/BSE framework, validating the use of the quantum fluctuations approach for calculating dynamic response functions.
• Scalability: By establishing that the two-time PA is equivalent to GW under HF-GKBA, the authors demonstrate that one can achieve the accuracy of GW approximations with the computational efficiency of the fluctuation approach.
• Memory Efficiency: The approach allows for the calculation of two-time correlation functions (essential for dynamic structure factors and response functions) without storing the full rank-four two-particle Green function tensor. This makes it feasible to simulate larger systems and longer timescales than previously possible with standard NEGF methods.
• Future Applications: The method is particularly promising for studying non-equilibrium dynamics in large-scale correlated systems (e.g., warm dense matter, cold atoms) where both accuracy and computational tractability are critical.