Neural networks for Bayesian quantum many-body magnetometry

TL;DR

This paper introduces neural networks that model quantum many-body sensor dynamics, enabling efficient Bayesian parameter estimation and surpassing standard quantum limit scaling in magnetometry.

Contribution

It presents a novel neural network approach to simulate quantum many-body dynamics for Bayesian inference, facilitating practical quantum sensing beyond traditional modeling limitations.

Findings

Neural networks accurately reproduce quantum many-body dynamics.

The method achieves parameter estimation beyond the standard quantum limit.

Applicable to XXZ models driven by magnetic fields.

Abstract

Entangled quantum many-body systems can be used as sensors that enable the estimation of parameters with a precision larger than that achievable with ensembles of individual quantum detectors. Typically, the parameter estimation strategy requires the microscopic modelling of the quantum many-body system, as well as a an accurate description of its dynamics. This entails a complexity that can hinder the applicability of Bayesian inference techniques. In this work we show how to circumvent these issues by using neural networks that faithfully reproduce the dynamics of quantum many-body sensors, thus allowing for an efficient Bayesian analysis. We exemplify with an XXZ model driven by magnetic fields, and show that our method is capable to yield an estimation of field parameters beyond the standard quantum limit scaling. Our work paves the way for the practical use of quantum many-body systems as black-box sensors exploiting quantum resources to improve precision estimation.