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 2103.10309quant-phcs.NAmath.NA
收藏

求解线性系统的更快速量子启发算法

Faster quantum-inspired algorithms for solving linear systems

Changpeng Shao, Ashley Montanaro

AI总结:

本文提出一种基于随机Kaczmarz方法的量子启发经典算法,将求解线性系统的复杂度从 $\widetilde{O}(κ_F^6 κ^6/ε^4)$ 改进至 $\widetilde{O}(κ_F^4 κ^2/ε^2)$,并证明在行稀疏矩阵情形下量子加速仅为二次方。

摘要:

我们建立了一种改进的经典算法,用于在与量子线性求解器所使用的QRAM类似的模型中求解线性系统。具体而言,对于线性系统 $A\x = \b$,我们证明存在一种经典算法,能够输出一个针对 $\x$ 的数据结构,允许对其元素进行采样和查询,其中 $\x$ 满足 $\\|\x - A^{+}\b\\|\leq ε\\|A^{+}\b\\|$。该输出可被视为量子线性求解器输出的经典对应物。我们算法的复杂度为 $\widetilde{O}(κ_F^4 κ^2/ε^2 )$,其中 $κ_F = \\|A\\|_F\\|A^{+}\\|$,$κ= \\|A\\|\\|A^{+}\\|$。这改进了此前的最佳算法[Gilyén, Song and Tang, arXiv:2009.07268],其复杂度为 $\widetilde{O}(κ_F^6 κ^6/ε^4)$。我们的算法基于随机Kaczmarz方法,该方法是随机梯度下降的一种特例。我们还发现,当 $A$ 是行稀疏矩阵时,该方法已经能在 $\widetilde{O}(κ_F^2)$ 时间内返回近似解 $\x$,而已知的最佳量子算法在 $A$ 存储于QRAM数据结构中时能在 $\widetilde{O}(κ_F)$ 时间内返回 $\ket{\x}$。因此,在假设能够访问QRAM且 $A$ 为行稀疏矩阵的条件下,基于当前量子算法的加速效果是二次方的。

英文摘要:

We establish an improved classical algorithm for solving linear systems in a model analogous to the QRAM that is used by quantum linear solvers. Precisely, for the linear system $A\x = \b$, we show that there is a classical algorithm that outputs a data structure for $\x$ allowing sampling and querying to the entries, where $\x$ is such that $\|\x - A^{+}\b\|\leq ε\|A^{+}\b\|$. This output can be viewed as a classical analogue to the output of quantum linear solvers. The complexity of our algorithm is $\widetilde{O}(κ_F^4 κ^2/ε^2 )$, where $κ_F = \|A\|_F\|A^{+}\|$ and $κ= \|A\|\|A^{+}\|$. This improves the previous best algorithm [Gily{é}n, Song and Tang, arXiv:2009.07268] of complexity $\widetilde{O}(κ_F^6 κ^6/ε^4)$. Our algorithm is based on the randomized Kaczmarz method, which is a particular case of stochastic gradient descent. We also find that when $A$ is row sparse, this method already returns an approximate solution $\x$ in time $\widetilde{O}(κ_F^2)$, while the best quantum algorithm known returns $\ket{\x}$ in time $\widetilde{O}(κ_F)$ when $A$ is stored in the QRAM data structure. As a result, assuming access to QRAM and if $A$ is row sparse, the speedup based on current quantum algorithms is quadratic.

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。

↑