IQPopt：在JAX中快速优化瞬时量子多项式电路
IQPopt: Fast optimization of instantaneous quantum polynomial circuits in JAX
AI总结：
IQPopt利用高效的经典模拟算法优化大规模瞬时量子多项式电路，通过Pauli-Z型可观测量高效描述目标函数，支持千级量子比特和百万级门操作的优化。
摘要：
IQPopt是一个用于在经典硬件上优化大规模瞬时量子多项式电路的软件包。通过利用高效的经典模拟算法估计期望值，可以优化具有数千个量子比特和数百万个门的电路，前提是相关的目标函数可以高效地以Pauli-Z型可观测量表示。由于从瞬时量子多项式电路采样被认为对经典计算机来说是困难的，这提供了一种在部署和在量子硬件上采样之前识别强大电路实例的方法，其中可能存在计算优势。该软件包利用JAX中的自动微分，可以通过访问图形处理单元等硬件加速器进行加速，并包含一个专用模块，可用于通过最大均值差异训练和评估量子生成模型。
英文摘要：
IQPopt is a software package designed to optimize large-scale instantaneous quantum polynomial circuits on classical hardware. By exploiting an efficient classical simulation algorithm for expectation value estimation, circuits with thousands of qubits and millions of gates can be optimized, provided the relevant objective function has an efficient description in terms of Pauli-Z type observables. Since sampling from instantaneous quantum polynomial circuits is widely believed to be hard for classical computers, this provides a method to identify powerful circuit instances before deployment and sampling on quantum hardware, where computational advantages may exist. The package leverages automatic differentiation in JAX, can be accelerated with access to hardware accelerators such as graphics processing units, and contains a dedicated module that can be used to train and evaluate quantum generative models via the maximum mean discrepancy.
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。
