Dynamical quantum maps for single-qubit gates under universal non-Markovian noise

TL;DR

This paper introduces a compact, experimentally accessible microscopic error model for single-qubit gates that accounts for non-Markovian noise, improving accuracy over standard models and aiding quantum error correction.

Contribution

The authors develop a noise model based solely on the noise power spectral density that explicitly includes non-Markovian effects, surpassing traditional depolarizing models.

Findings

Model predictions tightly lower-bound experimental gate errors

Depolarizing models tend to overestimate errors

Non-Markovian noise significantly impacts quantum gate fidelity

Abstract

Noise is both ubiquitous and generally deleterious in settings where precision is required. This is especially true in the quantum technology sector where system utility typically decays rapidly under its influence. Understanding the noise in quantum devices is thus a prerequisite for efficient strategies to mitigate or even eliminate its harmful effects. However, this requires resources that are often prohibitive, such that the typically-used noise models rely on simplifications that sometimes depart from experimental reality. Here we derive a compact microscopic error model for single-qubit gates that only requires a single experimental input - the noise power spectral density. Our model goes beyond standard depolarizing or Pauli-twirled noise models, explicitly including non-Clifford and non-Markovian contributions to the dynamical error map. We gauge our predictions for experimentally relevant metrics against established characterization techniques run on a trapped-ion quantum computer. In particular, we find that experimental estimates of average gate errors measured through randomized benchmarking and reconstructed via quantum process tomography are tightly lower-bounded by our analytical estimates, while the depolarizing model overestimates the gate error. Our noise modeling including non-Markovian contributions can be readily applied to established frameworks such as dynamical decoupling and dynamically-corrected gates, or to provide more realistic thresholds for quantum error correction.