Average circuit eigenvalue sampling on NISQ devices

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pelaez, Emilio, Omole, Victory, Gokhale, Pranav, Rines, Rich, Smith, Kaitlin N., Perlin, Michael A., Hashim, Akel
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916167020445696
author Pelaez, Emilio
Omole, Victory
Gokhale, Pranav
Rines, Rich
Smith, Kaitlin N.
Perlin, Michael A.
Hashim, Akel
author_facet Pelaez, Emilio
Omole, Victory
Gokhale, Pranav
Rines, Rich
Smith, Kaitlin N.
Perlin, Michael A.
Hashim, Akel
contents Average circuit eigenvalue sampling (ACES) was introduced by Flammia in arXiv:2108.05803 as a protocol to characterize the Pauli error channels of individual gates across the device simultaneously. The original paper posed using ACES to characterize near-term devices as an open problem. This work advances in this direction by presenting a full implementation of ACES for real devices and deploying it to Superstaq arXiv:2309.05157, along with a device-tailored resource estimation obtained through simulations and experiments. Our simulations show that ACES is able to estimate one- and two-qubit non-uniform Pauli error channels to an average eigenvalue absolute error of under $0.003$ and total variation distance of under 0.001 between simulated and reconstructed probability distributions over Pauli errors with $10^5$ shots per circuit using 5 circuits of depth 14. The question of estimating general error channels through twirling techniques in real devices remains open, as it is dependent on a device's native gates, but simulations with the Clifford set show results in agreement with reported hardware data. Experimental results on IBM's Algiers and Osaka devices are presented, where we characterize their error channels as Pauli channels without twirling.
format Preprint
id arxiv_https___arxiv_org_abs_2403_12857
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Average circuit eigenvalue sampling on NISQ devices
Pelaez, Emilio
Omole, Victory
Gokhale, Pranav
Rines, Rich
Smith, Kaitlin N.
Perlin, Michael A.
Hashim, Akel
Quantum Physics
Average circuit eigenvalue sampling (ACES) was introduced by Flammia in arXiv:2108.05803 as a protocol to characterize the Pauli error channels of individual gates across the device simultaneously. The original paper posed using ACES to characterize near-term devices as an open problem. This work advances in this direction by presenting a full implementation of ACES for real devices and deploying it to Superstaq arXiv:2309.05157, along with a device-tailored resource estimation obtained through simulations and experiments. Our simulations show that ACES is able to estimate one- and two-qubit non-uniform Pauli error channels to an average eigenvalue absolute error of under $0.003$ and total variation distance of under 0.001 between simulated and reconstructed probability distributions over Pauli errors with $10^5$ shots per circuit using 5 circuits of depth 14. The question of estimating general error channels through twirling techniques in real devices remains open, as it is dependent on a device's native gates, but simulations with the Clifford set show results in agreement with reported hardware data. Experimental results on IBM's Algiers and Osaka devices are presented, where we characterize their error channels as Pauli channels without twirling.
title Average circuit eigenvalue sampling on NISQ devices
topic Quantum Physics
url https://arxiv.org/abs/2403.12857