Predicting non-Markovian superconducting qubit dynamics from tomographic reconstruction

TL;DR

This paper demonstrates that a simple phenomenological model, the post-Markovian master equation, can accurately predict non-Markovian noise in superconducting qubits by using tomographic data from an IBM quantum processor, outperforming standard models.

Contribution

The study shows that the PMME model, built from experimental data, effectively captures and predicts non-Markovian qubit dynamics, providing insights into cross-talk and non-Markovianity measures.

Findings

PMME accurately predicts non-Markovian qubit dynamics.

Model outperforms standard Markovian master equation.

Extracts information on cross-talk and non-Markovianity.

Abstract

Non-Markovian noise presents a particularly relevant challenge in understanding and combating decoherence in quantum computers, yet is challenging to capture in terms of simple models. Here we show that a simple phenomenological dynamical model known as the post-Markovian master equation (PMME) accurately captures and predicts non-Markovian noise in a superconducting qubit system. The PMME is constructed using experimentally measured state dynamics of an IBM Quantum Experience cloud-based quantum processor, and the model thus constructed successfully predicts the non-Markovian dynamics observed in later experiments. The model also allows the extraction of information about cross-talk and measures of non-Markovianity. We demonstrate definitively that the PMME model predicts subsequent dynamics of the processor better than the standard Markovian master equation.