arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.09601quant-ph
收藏

简化量子权重降低算法及其最优界

Simplified Quantum Weight Reduction with Optimal Bounds

Foxconn Research(富士康研究);Department of Computer Science and Technology, Tsinghua University(清华大学计算机科学与技术系);Department of Physics, University of California San Diego(加州大学圣地亚哥分校物理系)

Min-Hsiu Hsieh, Xingjian Li, Ting-Chun Lin

AI总结:

本文提出一种结合几何与锥形技术的简化量子权重降低方法,生成低权重码,超越平方根距离障碍并饱和BPT界,同时改进容错逻辑测量。

摘要:

量子权重降低是将具有大校验权重的量子码转换为具有小校验权重的量子码的任务。低权重码对于在物理硬件上实现量子纠错至关重要,因为高权重测量无法可靠执行。权重降低也是重要的理论工具,可能与量子PCP猜想相关。我们提出了一种新的量子权重降低程序,结合了几何洞察与锥形技术,简化了Hastings先前的方法,同时实现了更好的参数。给定任意权重为$w$的$[[n,k,d]]$量子码,我们的方法生成参数为$[[O(n w^2 \log w), k, \Omega(d w)]]$、校验权重为5、量子比特权重为6的码。当应用于随机稠密CSS码时,我们的程序产生了超越平方根距离障碍的显式量子码,实现参数$[[n, \tilde O(n^{1/3}), \tilde \Omega(n^{2/3})]]$。此外,这些码允许三维嵌入,饱和了Bravyi-Poulin-Terhal(BPT)界。作为进一步应用,我们的权重降低技术通过减少辅助量子比特的数量,改进了容错逻辑运算符测量。

英文摘要:

Quantum weight reduction is the task of transforming a quantum code with large check weight into one with small check weight. This problem is important in practice, since low-weight measurements are necessary for reliable implementations of quantum error correction on physical hardware. It is also important theoretically, as it can be used to construct constant-locality versions of quantum locally testable codes, which may be relevant to the quantum PCP conjecture. We give a streamlined geometric procedure for quantum weight reduction based entirely on coning and treating X and Z checks symmetrically, which simplifies Hastings' previous approach. In particular, given an arbitrary $[[n,k,d]]$ quantum code with weight $w$, our method produces a code with parameters $[[O(n w^2 \log w), k, Ω(d w)]]$, check weight $5$, and qubit weight $6$; these bounds are optimal or close to optimal within the current geometric framework. Applied to random dense CSS codes, our procedure yields explicit quantum codes surpassing the square-root distance barrier, with parameters $[[n, \tilde Ω(n^{1/3}), \tilde Ω(n^{2/3})]]$. These codes also admit a three-dimensional embedding that saturates the Bravyi-Poulin-Terhal (BPT) bound, recovering the layer-code result. Our weight reduction technique also improves fault-tolerant logical-operator measurements by reducing the required number of ancilla qubits.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑