论文标题

来自恒定数量组件的耐故障量子

Fault-tolerant qubit from a constant number of components

论文作者

Wan, Kianna, Choi, Soonwon, Kim, Isaac H., Shutty, Noah, Hayden, Patrick

论文摘要

现在,由于多种技术中的栅极错误率现在低于易于故障量子计算所需的阈值,因此有用的量子计算的主要剩余障碍是缩放,这一挑战大大增加了量子误差校正本身所施加的巨大开销。我们提出了一种容忍故障的量子计算方案,尽管可以从少数实验组件中组装出来,该方案可能会大大减少与构建大规模易耐故障量子计算机相关的工程挑战。假设记忆误差可忽略不计,我们的外观噪声的阈值为0.39%。在存在内存错误的情况下,逻辑错误率以$ \ sqrt {t/τ}​​ $呈指数衰减,其中$ t $是内存相干时间,而$τ$是基本门的时间表。我们的方法基于一种新的程序,用于使用单个主动控制的值和两对延迟线来耐断层制备三维群集状态。尽管电路级误差可能会传播到高重量误差,但此误差对准备的状态的影响始终等于恒定误差。我们描述了如何使用量子光子和声音系统中的现有技术实现必要的门。随着仅几个组件的持续改进,我们希望这些系统将成为有希望的候选人,以表现出易于故障的量子计算,并具有相对适度的实验努力。

With gate error rates in multiple technologies now below the threshold required for fault-tolerant quantum computation, the major remaining obstacle to useful quantum computation is scaling, a challenge greatly amplified by the huge overhead imposed by quantum error correction itself. We propose a fault-tolerant quantum computing scheme that can nonetheless be assembled from a small number of experimental components, potentially dramatically reducing the engineering challenges associated with building a large-scale fault-tolerant quantum computer. Our scheme has a threshold of 0.39% for depolarising noise, assuming that memory errors are negligible. In the presence of memory errors, the logical error rate decays exponentially with $\sqrt{T/τ}$, where $T$ is the memory coherence time and $τ$ is the timescale for elementary gates. Our approach is based on a novel procedure for fault-tolerantly preparing three-dimensional cluster states using a single actively controlled qubit and a pair of delay lines. Although a circuit-level error may propagate to a high-weight error, the effect of this error on the prepared state is always equivalent to that of a constant-weight error. We describe how the requisite gates can be implemented using existing technologies in quantum photonic and phononic systems. With continued improvements in only a few components, we expect these systems to be promising candidates for demonstrating fault-tolerant quantum computation with a comparatively modest experimental effort.

扫码加入交流群

加入微信交流群

微信交流群二维码

扫码加入学术交流群,获取更多资源