Generalized quantum subspace expansion

TL;DR

This paper introduces a generalized quantum subspace expansion method that effectively mitigates various types of errors in quantum computers without prior noise information, enhancing error suppression in practical quantum computing scenarios.

Contribution

It presents a novel, hardware-friendly quantum error mitigation technique that handles stochastic, coherent, and algorithmic errors by exploiting an extended subspace, improving error suppression capabilities.

Findings

Suppresses errors by orders of magnitude in simulations

Handles stochastic, coherent, and algorithmic errors effectively

Overcomes limitations of previous error-agnostic QEM methods

Abstract

One of the major challenges for erroneous quantum computers is undoubtedly the control over the effect of noise. Considering the rapid growth of available quantum resources that are not fully fault-tolerant, it is crucial to develop practical hardware-friendly quantum error mitigation (QEM) techniques to suppress unwanted errors. Here, we propose a novel generalized quantum subspace expansion method which can handle stochastic, coherent, and algorithmic errors in quantum computers. By fully exploiting the substantially extended subspace, we can efficiently mitigate the noise present in the spectra of a given Hamiltonian, without relying on any information of noise. The performance of our method is discussed under two highly practical setups: the quantum subspaces are mainly spanned by powers of the noisy state $\rho^m$ and a set of error-boosted states, respectively. We numerically demonstrate in both situations that we can suppress errors by orders of magnitude, and show that out protocol inherits the advantages of previous error-agnostic QEM techniques as well as overcoming their drawbacks.