Beyond Quantum Noise Spectroscopy: modelling and mitigating noise with quantum feature engineering

TL;DR

This paper introduces a deep learning framework utilizing quantum feature engineering for comprehensive quantum system characterization and noise mitigation, enabling better control and understanding of quantum devices despite environmental noise.

Contribution

It presents a novel deep learning approach with quantum features for quantum system characterization, including tools for noise spectrum extraction and mitigation.

Findings

The framework accurately models quantum noise and control.

It can extract noise power spectra from data.

Demonstrates improved quantum system management.

Abstract

The ability to use quantum technology to achieve useful tasks, be they scientific or industry related, boils down to precise quantum control. In general it is difficult to assess a proposed solution due to the difficulties in characterising the quantum system or device. These arise because of the impossibility to characterise certain components in situ, and are exacerbated by noise induced by the environment and active controls. Here we present a general purpose characterisation and control solution making use of a novel deep learning framework composed of quantum features. We provide the framework, sample data sets, trained models, and their performance metrics. In addition, we demonstrate how the trained model can be used to extract conventional indicators, such as noise power spectra.