Estimating the coherence of noise in mid-scale quantum systems

TL;DR

This paper introduces an efficient protocol to estimate the coherence of noise in mid-scale quantum systems, providing practical tools for assessing and improving quantum device performance.

Contribution

It presents a new method to efficiently estimate the unitarity of noise in quantum devices, scalable to mid-sized systems, with experimental validation on a 5-qubit quantum computer.

Findings

Successfully estimated unitarity on a 5-qubit device

Simulated unitarity estimation for up to 10 qubits

Discussed potential for scaling to larger systems

Abstract

While the power of quantum computers is commonly acknowledged to rise exponentially, it is often overlooked that the complexity of quantum noise mechanisms generally grows much faster. In particular, quantifying whether the instructions on a quantum processor are close to being unitary has important consequences concerning error rates, e.g., for the confidence in their estimation, the ability to mitigate them efficiently, or their relation to fault-tolerance thresholds in error correction. However, the complexity of estimating the coherence, or unitarity, of noise generally scales exponentially in system size. Here, we obtain an upper bound on the average unitarity of Pauli noise and develop a protocol allowing us to estimate the average unitarity of operations in a digital quantum device efficiently and feasibly for mid-size quantum systems. We demonstrate our results through both experimental execution on IQM Spark (TM), a 5-qubit superconducting quantum computer, and in simulation with up to 10 qubits, discussing the prospects for extending our technique to arbitrary scales.