Sample-efficient quantum error mitigation via classical learning surrogates

Wei-You Liao, Ge Yan, Yujin Song, Tian-Ci Tian, Wei-Ming Zhu, De-Tao Jiang, Yuxuan Du, He-Liang Huang · 2025

Read Sample-efficient quantum error mitigation via classical learning surrogates by Wei-You Liao, Ge Yan, Yujin Song, Tian-Ci Tian, Wei-Ming Zhu, De-Tao Jiang, Yuxuan Du, He-Liang Huang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-07092.txt.

Description

Wei-You Liao and seven co-authors cut the measurement cost of quantum error mitigation, the set of tricks that clean up results from noisy quantum processors. Mitigation avoids the huge overhead of full error correction, but tools like zero-noise extrapolation must remeasure every circuit variant. The cost grows linearly when a whole family of circuits is parameterized by classic inputs. Their surrogate-enabled version moves the extrapolation onto classic computers.

A classic learning surrogate is trained to imitate the noisy device, after which the extrapolation runs entirely on the surrogate. The measurement overhead becomes constant for an entire circuit family rather than growing with its size. Theory indicates accuracy comparable to standard extrapolation in many useful cases, and numerical tests on systems up to one hundred qubits, covering ground-state energies and quantum metrology, confirm the method works.

Catalog record

CREATORWei-You Liao, Ge Yan, Yujin Song, Tian-Ci Tian, Wei-Ming Zhu, De-Tao Jiang, Yuxuan Du, He-Liang Huang
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE83 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-07092

Browse the full library · About the Tower of Babel