Measurement Noise Mitigation in a Quantum Computer Using Image Intensity Filters

TL;DR

This paper introduces a novel image contrast filter method to reduce measurement errors in quantum computers, outperforming existing techniques like M3 across different platforms such as IBM-Q and IonQ.

Contribution

The paper presents a new image contrast filter approach for measurement noise mitigation that does not rely on linear equations, demonstrating superior performance and platform independence.

Findings

Outperforms M3 measurement mitigation in experiments

Effective on both IBM-Q and IonQ quantum platforms

Provides open-source code and detailed documentation

Abstract

We propose a method to mitigate measurement errors in the distribution counts of a Quantum computer using image contrast filters. This work is similar to the method described by Gambetta and colleagues in [1]; however our technique does not use a linear system of equations, but an image contrast filter to mitigate the measurement noise. Furthermore this method is demonstrated against the same set of experiments described in the matrix-free measurement mitigation (M3) library from Qiskit from which [1] is based upon. Our results show our method outperforming M3 by a wide margin in all experiments on IBM-Q. Furthermore, our method is platform agnostic; we demonstrate this by running some experiments on the IonQ cloud with similar results. Finally, we provide results, documentation and detailed test and source code for further investigation.