Skip to main content
QUICK REVIEW

[论文解读] Efficient Preparation of Graph States using the Quotient-Augmented Strong Split Tree

Nicholas Connolly, S. Nishio|arXiv (Cornell University)|Mar 25, 2026
Quantum Computing Algorithms and Architecture被引用 0
一句话总结

该论文提出一个 quotient-augmented strong split tree (QASST) 框架,用于对距离保持图(distance-hereditary graphs)的 LC 轨道进行分类,并引入 split-fuse 和 greedy 策略来实现可扩展的图态制备,在 DH 图上的 CZ 门、时间步骤和辅助量子比特方面实现线性扩展。

ABSTRACT

Graph states are a key resource for measurement-based quantum computation and quantum networking, but state-preparation costs limit their practical use. Graph states related by local complement (LC) operations are equivalent up to single-qubit Clifford gates; one may reduce entangling resources by preparing a favorable LC-equivalent representative. However, exhaustive optimization over the LC orbit is not scalable. We address this problem using the split decomposition and its quotient-augmented strong split tree (QASST). For several families of distance-hereditary (DH) graphs, we use the QASST to characterize LC orbits and identify representatives with reduced controlled-Z count or preparation circuit depth. We also introduce a split-fuse construction for arbitrary DH graph states, achieving linear scaling with respect to entangling gates, time steps, and auxiliary qubits. Beyond the DH setting, we discuss a generalized divide-and-conquer split-fuse strategy and a simple greedy heuristic for generic graphs based on triangle enumeration. Together, these methods outperform direct implementations on sufficiently large graphs, providing a scalable alternative to brute-force optimization.

研究动机与目标

  • 通过利用 LC 等价性和结构图分解,激励高效制备大型图态。
  • 使用 QASST 表征距离保持图的 LC 轨道,以识别需要更少 CZ 门或更浅电路的 LC 代表。
  • 引入 split-fuse 构造,通过从商图组装目标 DH 图态,使 CZ 门、时间步和辅助量子比特的扩展达到线性规模。
  • 在通用图上推广该方法,提出广义 split-fuse 策略和基于贪心三角形的启发式方法。
  • 提供实际工具(如 Python 脚本)和显式的 LC-转移序列,以实现图态制备的可扩展优化。

提出的方法

  • 使用 quotient-augmented strong split tree (QASST) 通过商图连接强分裂树来描述图。
  • 通过分析 QASST 框架内的局部等价商图来对距离保持图的 LC 轨道进行分类。
  • 对 DH 家族(如完全二部图、完全多部图、团-星图、中继图)推导具有最小边数或最大顶点度数的 LC-轨道代表。
  • 引入 split-fuse:一种制备协议,通过 Type-II 融合和线性时间计算的分解将商图状态拼接成目标 DH 图态。
  • 为通用图提出广义 split-fuse,适用于星形/完全商商图;在处理素商商图时应用贪心三角形枚举启发式。
  • 提供基于三角形的贪心优化(Algorithm 1)以在不进行完整 LC-轨道搜索的情况下减少边数。
Figure 2 : (a) The effect of local complement on a graph $G$ with respect to vertex $1\in V(G)$ (green vertex). The edges between the neighbors of 1 (yellow vertices) are complemented: existing edges are deleted, and missing edges are added. (b) The LC orbit of the complete graph on three vertices,
Figure 2 : (a) The effect of local complement on a graph $G$ with respect to vertex $1\in V(G)$ (green vertex). The edges between the neighbors of 1 (yellow vertices) are complemented: existing edges are deleted, and missing edges are added. (b) The LC orbit of the complete graph on three vertices,

实验结果

研究问题

  • RQ1如何在不进行穷尽的 LC 枚举情况下高效表征距离保持图的 LC 轨道?
  • RQ2QASST 框架是否能为 DH 图族(如完全二部、完全多部、团-星)推导出解析的 LC-最优代表?
  • RQ3是否可通过 split-fuse 构造在 CZ 门、时间步和辅助量子比特方面实现线性扩展的大图态制备?
  • RQ4在 DH 结构缺失时,如何使用广义 split-fuse 和简单启发式处理通用图?
  • RQ5哪些实际工具(如算法、脚本)能够在 LC 等价类中实现图态制备的可扩展优化?

主要发现

QASST-sym.countQ2 sym.counttotal|E(G)|Δ(G)Trans. from K_{n,m}
star-center1star-center11nmmax{n,m}id
star-center1complete11nm+ m(m-1)/2n+m-1c_{i^{n}}
complete1star-center11nm+ n(n-1)/2n+m-1c_{i^{m}}
star-spokencomplete1nnm-1 + m(m-1)/2n+m-1c_{i^{n}}
complete1star-spokemmn+m-1 + n(n-1)/2n+m-1c_{i^{m}}
star-spokenstar-spokemnmn+m-1max{n,m}c_{i^{n}}∘ c_{i^{m}}⨀ c_{i^{n}}
  • 对于完全二部图 K_{n,m},LC 轨道大小为 nm+n+m+3,轨道中最小边数为 n+m-1(由 binary-star 实现),最大度 Δ(G)=max{n,m}。
  • 基于 QASST 的分类扩展到完全多部图和团-星图,能够进行明确的 LC-轨道分析并为这些 DH 家族识别最优代表(详见附录 D)。
  • split-fuse 构造通过从商图拼接而来实现 DH 图态的线性扩展,在 CZ 门、时间步和辅助量子比特方面具有线性缩放;它利用线性时间的分解和 Type-II 融合。
  • 超越 DH 图,广义 split-fuse 策略结合星形/完全商商图使得非 DH 图也可处理;配以简单的贪心三角形枚举启发式以减少边数。
  • 贪心算法 1(Triangle-based greedy)在最坏情况下的复杂度为 O(|E(G)|^{1.5} + |V(G)|^{3}),且不需要事先的 LC-轨道知识,即可高效降低边数。
  • 该方法在结构性 LC-轨道洞见和可扩展制备方法上共同提供优势,能在足够大图上超越穷举优化效果。
Figure 3 : The split decompositions for the three special families of DH graph that we consider. $K_{n,m}$ splits into two quotient graphs, while $K_{n_{1},\cdots,n_{k}}$ and $CS^{r}_{n_{1},\cdots,n_{k}}$ always split into $k+1$ quotient graphs. We adopt the convention of labeling the central quotie
Figure 3 : The split decompositions for the three special families of DH graph that we consider. $K_{n,m}$ splits into two quotient graphs, while $K_{n_{1},\cdots,n_{k}}$ and $CS^{r}_{n_{1},\cdots,n_{k}}$ always split into $k+1$ quotient graphs. We adopt the convention of labeling the central quotie

更好的研究,从现在开始

从论文设计到论文写作,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。