Dual Map Framework for Noise Characterization of Quantum Computers

TL;DR

This paper introduces a dual map framework and a local noise reconstruction method (MATEN) to efficiently characterize and benchmark error channels in quantum computers, validated through simulations and experiments on Rigetti's quantum device.

Contribution

It presents a novel dual map framework and the MATEN method for local noise characterization, improving understanding of quantum device errors.

Findings

Successfully reconstructs local noise channels on quantum hardware.

Accurately predicts measurement outcomes in QAOA circuits.

Validates the approach with numerical simulations and real device experiments.

Abstract

In order to understand the capabilities and limitations of quantum computers, it is necessary to develop methods that efficiently characterize and benchmark error channels present on these devices. In this paper, we present a method that faithfully reconstructs a marginal (local) approximation of the effective noise (MATEN) channel, that acts as a single layer at the end of the circuit. We first introduce a dual map framework that allows us to analytically derive expectation values of observables with respect to noisy circuits. These findings are supported by numerical simulations of the quantum approximate optimization algorithm (QAOA) that also justify the MATEN, even in the presence of non-local errors that occur during a circuit. Finally, we demonstrate the performance of the method on Rigetti's Aspen-9 quantum computer for QAOA circuits up to six qubits, successfully predicting the observed measurements on a majority of the qubits.