arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.07332quant-ph
收藏

通过分段QSVT进行量子态制备

Quantum state preparation via piecewise QSVT

Oliver O'Brien, Christoph Sünderhauf

AI总结:

本文提出利用分段量子奇异值变换和分段线性对角块编码高效制备振幅可由分段多项式近似的量子态,并将B样条窗口态制备成本较Kaiser窗口降低50倍。

摘要:

高效的量子态制备对于实现高效的量子算法至关重要。虽然已有若干低成本量子态制备技术,但本工作进一步支持其振幅可由分段多项式很好近似的新类型量子态。我们展示了如何利用分段量子奇异值变换以及一种新的分段线性对角块编码来高效制备此类量子态。我们以 $x^α|x\rangle$ 和 $\log x|x\rangle$ 作为显式示例加以说明。此外,我们的技术通过高效制备B样条窗口态,降低了窗口增强量子相位估计的成本。我们证明该窗口态所需的Toffoli门数量比最先进的Kaiser窗口态少50倍,并且表明对于量子相位估计,B样条窗口在尾部概率上复现了Kaiser窗口的指数级降低效果。

英文摘要:

Efficient state preparation is essential for implementing efficient quantum algorithms. Whilst several techniques for low-cost state preparation exist, this work facilitates further classes of states, whose amplitudes are well approximated by piecewise polynomials. We show how such states can be efficiently prepared using a piecewise Quantum Singular Value Transformation along with a new piecewise linear diagonal block encoding. We illustrate this with the explicit examples of $x^α|x\rangle$ and $\log x|x\rangle$. Further, our technique reduces the cost of window boosted Quantum Phase Estimation by efficiently preparing the B-spline window state. We demonstrate this window state requires 50 times fewer Toffolis to prepare than the state-of-the-art Kaiser window state, and we show that the B-spline window replicates the Kaiser window's exponential reduction in tail probability for QPE.

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。

↑