# Noise Detection with Spectator Qubits and Quantum Feature Engineering

arXiv: https://arxiv.org/abs/2103.13018 · HTML: https://mlsanity.com/paper/2103.13018/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 Mar 2021.

## Built on by (most cited first, 4 of 4)

- [On the Interpretability of Quantum Neural Networks](https://mlsanity.com/paper/2308.11098/) (2023)
- [Singularity-free dynamical invariants-based quantum control](https://mlsanity.com/paper/2510.15340/) (2025)
- [Model-free Distortion Canceling and Control of Quantum Devices](https://mlsanity.com/paper/2407.09877/) (2024)
- [Bayesian quantum sensing using graybox machine learning](https://mlsanity.com/paper/2601.17465/) (2026)

## Abstract (arXiv)

Designing optimal control pulses that drive a noisy qubit to a target state is a challenging and crucial task for quantum engineering. In a situation where the properties of the quantum noise affecting the system are dynamic, a periodic characterization procedure is essential to ensure the models are updated. As a result, the operation of the qubit is disrupted frequently. In this paper, we propose a protocol that addresses this challenge by making use of a spectator qubit to monitor the noise in real-time. We develop a machine-learning-based quantum feature engineering approach for designing the protocol. The complexity of the protocol is front-loaded in a characterization phase, which allow real-time execution during the quantum computations. We present the results of numerical simulations that showcase the favorable performance of the protocol.