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:cond-mat/0509144 · Physical Review A

Exact BCS stochastic schemes for a time-dependent many-body fermionic system

A. Montina, Yvan Castin

Published Jan 20, 2006• 13 citations
arXivHTMLPDF

Abstract

A common approximate way to study the time evolution of a superfluid Fermi gas is to consider the mean-field time-dependent BCS equations. We show that, by inclusion of an appropriate fluctuating term in these mean-field equations, the exact quantum evolution of the Fermi gas is recovered, after an average over all possible realizations of the noise term: this leads to an exact quantum Monte Carlo technique for the real-time evolution of an interacting fermionic system. In practice, we derive the general conditions on the noise term, ensuring that the exact dynamics is reproduced, and we explore several explicit realizations of this noise term, trying to minimize the statistical spreading of the quantum Monte Carlo simulations. The convergence issue is discussed in detail, with upper bounds obtained on the statistical spreading, and the relation between the stochastic and the mean-field equations is analyzed. Finally, we apply the method to a simple two-site model, to exemplify that the simulations can display effects out of reach of the mean-field approximation.

DOI
10.1103/physreva.73.013618

Related papers

Exact Stochastic Mean-Field Approach to the Fermionic Many-Body Problem

Olivier Juillet, Ph. Chomaz

We investigate a reformulation of the dynamics of interacting fermion systems in terms of a stochastic extension of time-dependent Hartree-Fock equations. From a path-integral representation of the evolution operator, we show that the exact N-body state can be interpreted as a coherent average over Slater determinants evolving in a random mean-field. The imaginary time propagation is also presented and gives a similar scheme which converges to the exact ground state. In addition, the growth of statistical errors is examined to show the stability of this stochastic formulation.

arXiv:nucl-th/0104085•Physical Review Letters•Mar 22, 2002•44 citations

Exact reformulation of the bosonic many-body problem in terms of stochastic wave functions: convergence issues

Iacopo Carusotto, Yvan Castin

There exist methods to reformulate in an exact way the many-body problem of interacting bosons in terms of the stochastic evolution of single particle wave functions. For one such reformulation, the so-called simple Fock scheme, we present an elementary derivation, much simpler than the original one. Furthermore, we show that two other schemes, based on coherent states of the matter field rather than on Fock states, lead to an infinite statistical uncertainty in the continuous time limit. The simple Fock scheme is therefore, up to now, the only one that was proved to lead to a convergent Monte Carlo simulation scheme at all times.

arXiv:cond-mat/0209135•Sep 5, 2002•1 citation

Exact and approximate many-body dynamics with stochastic one-body density matrix evolution

Denis Lacroix A

We show that the dynamics of interacting fermions can be exactly replaced by a quantum jump theory in the many-body density matrix space. In this theory, jumps occur between densities formed of pairs of Slater determinants, ${D}_{\mathit{ab}}=|{\ensuremath{\Phi}}_{a}\ensuremath{\rangle}\ensuremath{\langle}{\ensuremath{\Phi}}_{b}|$, where each state evolves according to the stochastic Schr\"odinger equation given by O. Juillet and Ph. Chomaz [Phys. Rev. Lett. 88, 142503 (2002)]. A stochastic Liouville-von Neumann equation is derived as well as the associated. Bogolyubov-Born-Green-Kirwood-Yvon hierarchy. Due to the specific form of the many-body density along the path, the presented theory is equivalent to a stochastic theory in one-body density matrix space, in which each density matrix evolves according to its own mean-field augmented by a one-body noise. Guided by the exact reformulation, a stochastic mean-field dynamics valid in the weak coupling approximation is proposed. This theory leads to an approximate treatment of two-body effects similar to the extended time-dependent Hartree-Fock scheme. In this stochastic mean-field dynamics, statistical mixing can be directly considered and jumps occur on a coarse-grained time scale. Accordingly, numerical effort is expected to be significantly reduced for applications.

arXiv:quant-ph/0407042•Physical Review C•Jun 30, 2005•4 citations

Sign-free stochastic mean-field approach to strongly correlated phases of ultracold fermions

Olivier Juillet

We propose a new projector quantum Monte-Carlo method to investigate the ground state of ultracold fermionic atoms modelled by a lattice Hamiltonian with on-site interaction. The many-body state is reconstructed from Slater determinants that randomly evolve in imaginary-time according to a stochastic mean-field motion. The dynamics prohibits the crossing of the exact nodal surface and no sign problem occurs in the Monte-Carlo estimate of observables. The method is applied to calculate ground-state energies and correlation functions of the repulsive two-dimensional Hubbard model. Numerical results for the unitary Fermi gas validate simulations with nodal constraints.

arXiv:cond-mat/0609063•New Journal of Physics•Jun 14, 2007•35 citations

Dynamics of Fluctuactions in the BCS-BEC Crossover in Superfluid Fermi Gases

H Vivas C

The BCS-BEC crossover phenomenon in a 3D attractive Fermi gas is revisited by using finite temperature Green functions formalism derived from the quantum path integral technique and Mean Field Approximation (MFA). Collective modes for weak and strong coupling particle-particle interaction g are obtained after recovering the dynamical coefficients in the nonlinear Time Dependent Ginzburg Landau (TDGL) equation, and numerically simulated under those particular states for the filling energy g ~ 0 and order parameter Δ( x , τ ) ~ Δ 0 . Explicit formulas for the first order average fluctuations ⟨η 2 q,ω ⟩ and the scaling law for the thermal energy associated to the non-linear collective modes are also reported.

Journal of Physics Conference Series•Dec 8, 2014•0 citations

Exact Monte Carlo time dynamics in many-body lattice quantum systems

Massimo Ostilli, Carlo Presilla A

On the basis of a Feynman–Kac-type formula involving Poisson stochastic processes, a Monte Carlo algorithm has recently been introduced, which describes exactly the real- or imaginary-time evolution of many-body lattice quantum systems. We extend this algorithm to the exact simulation of time-dependent correlation functions. The techniques generally employed in Monte Carlo simulations to control fluctuations, namely reconfigurations and importance sampling, are adapted to the present algorithm and their validity is rigorously proved. We complete the analysis by several examples for the hard-core boson Hubbard model and for the Heisenberg model.

arXiv:cond-mat/0407758•Journal of Physics A Mathematical and General•Dec 16, 2004•11 citations
A common approximate way to study the time evolution of a superfluid Fermi gas is to consider the mean-field time-dependent BCS equations. We show that, by inclusion of an appropriate fluctuating term in these mean-field equations, the exact quantum evolution of the Fermi gas is recovered, after an average over all possible realizations of the noise term: this leads to an exact quantum Monte Carlo technique for the real-time evolution of an interacting fermionic system. In practice, we derive the general conditions on the noise term, ensuring that the exact dynamics is reproduced, and we explore several explicit realizations of this noise term, trying to minimize the statistical spreading of the quantum Monte Carlo simulations. The convergence issue is discussed in detail, with upper bounds obtained on the statistical spreading, and the relation between the stochastic and the mean-field equations is analyzed. Finally, we apply the method to a simple two-site model, to exemplify that the simulations can display effects out of reach of the mean-field approximation.
We investigate a reformulation of the dynamics of interacting fermion systems in terms of a stochastic extension of time-dependent Hartree-Fock equations. From a path-integral representation of the evolution operator, we show that the exact N-body state can be interpreted as a coherent average over Slater determinants evolving in a random mean-field. The imaginary time propagation is also presented and gives a similar scheme which converges to the exact ground state. In addition, the growth of statistical errors is examined to show the stability of this stochastic formulation.
There exist methods to reformulate in an exact way the many-body problem of interacting bosons in terms of the stochastic evolution of single particle wave functions. For one such reformulation, the so-called simple Fock scheme, we present an elementary derivation, much simpler than the original one. Furthermore, we show that two other schemes, based on coherent states of the matter field rather than on Fock states, lead to an infinite statistical uncertainty in the continuous time limit. The simple Fock scheme is therefore, up to now, the only one that was proved to lead to a convergent Monte Carlo simulation scheme at all times.
We show that the dynamics of interacting fermions can be exactly replaced by a quantum jump theory in the many-body density matrix space. In this theory, jumps occur between densities formed of pairs of Slater determinants, ${D}_{\mathit{ab}}=|{\ensuremath{\Phi}}_{a}\ensuremath{\rangle}\ensuremath{\langle}{\ensuremath{\Phi}}_{b}|$, where each state evolves according to the stochastic Schr\"odinger equation given by O. Juillet and Ph. Chomaz [Phys. Rev. Lett. 88, 142503 (2002)]. A stochastic Liouville-von Neumann equation is derived as well as the associated. Bogolyubov-Born-Green-Kirwood-Yvon hierarchy. Due to the specific form of the many-body density along the path, the presented theory is equivalent to a stochastic theory in one-body density matrix space, in which each density matrix evolves according to its own mean-field augmented by a one-body noise. Guided by the exact reformulation, a stochastic mean-field dynamics valid in the weak coupling approximation is proposed. This theory leads to an approximate treatment of two-body effects similar to the extended time-dependent Hartree-Fock scheme. In this stochastic mean-field dynamics, statistical mixing can be directly considered and jumps occur on a coarse-grained time scale. Accordingly, numerical effort is expected to be significantly reduced for applications.
We propose a new projector quantum Monte-Carlo method to investigate the ground state of ultracold fermionic atoms modelled by a lattice Hamiltonian with on-site interaction. The many-body state is reconstructed from Slater determinants that randomly evolve in imaginary-time according to a stochastic mean-field motion. The dynamics prohibits the crossing of the exact nodal surface and no sign problem occurs in the Monte-Carlo estimate of observables. The method is applied to calculate ground-state energies and correlation functions of the repulsive two-dimensional Hubbard model. Numerical results for the unitary Fermi gas validate simulations with nodal constraints.
The BCS-BEC crossover phenomenon in a 3D attractive Fermi gas is revisited by using finite temperature Green functions formalism derived from the quantum path integral technique and Mean Field Approximation (MFA). Collective modes for weak and strong coupling particle-particle interaction g are obtained after recovering the dynamical coefficients in the nonlinear Time Dependent Ginzburg Landau (TDGL) equation, and numerically simulated under those particular states for the filling energy g ~ 0 and order parameter Δ( x , τ ) ~ Δ 0 . Explicit formulas for the first order average fluctuations ⟨η 2 q,ω ⟩ and the scaling law for the thermal energy associated to the non-linear collective modes are also reported.
On the basis of a Feynman–Kac-type formula involving Poisson stochastic processes, a Monte Carlo algorithm has recently been introduced, which describes exactly the real- or imaginary-time evolution of many-body lattice quantum systems. We extend this algorithm to the exact simulation of time-dependent correlation functions. The techniques generally employed in Monte Carlo simulations to control fluctuations, namely reconfigurations and importance sampling, are adapted to the present algorithm and their validity is rigorously proved. We complete the analysis by several examples for the hard-core boson Hubbard model and for the Heisenberg model.