Overdispersion in gate tomography: Experiments and continuous, two-scale random walk model on the Bloch sphere

What is it about?

This article presents a new approach to modelling the noise of so-called noisy intermediate-scale quantum computers. Simply put we superimpose a base random walk model by a second random walk at the scale of multiple readouts to account for the aggregation of noise effects over time. We calibrate our two-scale random walk model with experiments carried out on real-world quantum processors.

Featured Image

Why is it important?

Realistic noise models are essential for the development of practical quantum algorithms as real-world quantum processors are still scarce, meaning that a large part of the development relies on quantum computing emulators. The noise models implemented in many of these emulators often do not reflect the complex noise behaviour observed in practice but fall back to simplified, e.g., one-level random walk models. This paper is a first step into developing more realistic noise models and, in the long run, enabling quantum computing emulation with realistic noise behaviour.