betaFeedback

WorksPhysics and Astronomy

Probing equilibrium properties with Gaussian states beyond mean-field

Abstract

Computing equilibrium properties of interacting fermionic systems is central to many problems in condensed matter physics, yet computationally challenging. Mean-field methods tackle this problem by restricting to a tractable class of states, which, however, often fail to capture genuine many-body correlations. We illustrate this in the context of Gaussian mean-field methods by minimizing the energy variance of Gaussian states, showing that even states with the narrowest energy distributions poorly reproduce the microcanonical ensemble. To overcome this limitation, we develop a computationally efficient method to construct linear superpositions of Gaussian states whose energy variance can be systematically decreased by adding more terms to the superposition. We provide complementary results for the canonical ensemble, for which we propose a sampling algorithm based on convex combinations of Gaussian states. We benchmark our methods using the tt–VV model and find that both methods greatly outperform mean-field calculations. In particular, we obtain the correct critical exponents as well as a significantly improved estimate for the critical temperature. As an independent result, we prove a quantum Berry–Esseen theorem for fermionic Gaussian states, which states that the local density of states converges to a Gaussian distribution in the thermodynamic limit without any assumptions about the spatial correlations of the state.

Paper details

vlnc:W_xanenggr
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
arXiv nonexclusive license
Usage

views · 0

downloads · 0

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

  • Physics and Astronomy

Sol, Astra

View statement
Looking for related papers…