Error-Mitigated Quantum Random Access Memory
arxiv(2024)
摘要
As an alternative to quantum error correction, quantum error mitigation
methods, including Zero-Noise Extrapolation (ZNE), have been proposed to
alleviate run-time errors in current noisy quantum devices. In this work, we
propose a modified version of ZNE that provides for a significant performance
enhancement on current noisy devices. Our modified ZNE method extrapolates to
zero-noise data by evaluating groups of noisy data obtained from noise-scaled
circuits and selecting extrapolation functions for each group with the
assistance of estimated noisy simulation results. To quantify enhancement in a
real-world quantum application, we embed our modified ZNE in Quantum Random
Access Memory (QRAM) - a memory system important for future quantum networks
and computers. Our new ZNE-enhanced QRAM designs are experimentally implemented
on a 27-qubit noisy superconducting quantum device, the results of which
demonstrate that with reasonable estimated simulation results, QRAM fidelity is
improved significantly relative to traditional ZNE usage. Our results
demonstrate the critical role the extrapolation function plays in ZNE -
judicious choice of that function on a per-measurement basis can make the
difference between a quantum application being functional or non-functional.
更多查看译文
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要