Machine Learning-Enhanced Characterisation of Structured Spectral Densities: Leveraging the Reaction Coordinate Mapping

TL;DR

This paper introduces a machine learning approach to analyze structured spectral densities in open quantum systems, using reaction coordinate mapping to simulate dynamics and neural networks to classify spectral features beyond weak-coupling regimes.

Contribution

It presents a novel method combining reaction coordinate mapping with neural networks to characterize structured spectral densities in open quantum systems.

Findings

Neural networks can classify the number of Lorentzian peaks in spectral densities.

The method accurately predicts the central frequencies of spectral peaks.

It extends analysis capabilities beyond weak-coupling regimes.

Abstract

Spectral densities encode essential information about system-environment interactions in open-quantum systems, playing a pivotal role in shaping the system's dynamics. In this work, we leverage machine learning techniques to reconstruct key environmental features, going beyond the weak-coupling regime by simulating the system's dynamics using the reaction coordinate mapping. For a dissipative spin-boson model with a structured spectral density expressed as a sum of Lorentzian peaks, we demonstrate that the time evolution of a system observable can be used by a neural network to classify the spectral density as comprising one, two, or three Lorentzian peaks and accurately predict their central frequency.