Collaborative Research: EAGER: QIA: Large Scale QAOA Quantum Simulator

合作研究：EAGER：QIA：大规模 QAOA 量子模拟器

• 批准号: 2122793
• 负责人: Ilya Safro
• 金额: $ 10万
• 依托单位: University of Delaware
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-10-01 至 2023-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2122793&HistoricalAwards=false
• 关键词: Collaborative; Research; EAGER; QIA; Large

The goal of this project is to develop a specialized Quantum Approximate Optimization Algorithm (QAOA) quantum circuit simulator. QAOA is the most studied quantum optimization algorithm and is considered to be the prime candidate for demonstrating quantum advantage. There is a worldwide race underway amongst top quantum information science researchers to find combinatorial optimization problems and their instances that run efficiently and faster on quantum devices rather than on classical computers. One of the critical bottlenecks is to find circuit parameters faster on a classical computer to accelerate variational quantum-classical frameworks. The expected improvements to the developed simulator will dramatically increase the speed of QAOA simulations by at least one order of magnitude and significantly speed up research done on finding optimal QAOA circuit parameters. As a result, it will help the realization of quantum advantage by US scientists in this highly competitive field of science.The technical goal of this project is to carry out computational and algorithmic investigations in various node elimination methods for tensor contraction in the development of a scalable quantum simulator. The plan is to build upon the success of the combinatorial scientific computing community in developing elimination algorithms for such tasks as minimizing the error and complexity in matrix computations and optimizing the time and space complexity in automatic differentiation, and recent developments in treewidth optimization algorithms. Additionally, in this project, emphasis will be placed on scaling up relevant graph algorithms to achieve acceptable time/quality trade-off for large-scale quantum simulators. Algorithmic and software products of this project will include a specialized QAOA quantum circuit open-source simulator equipped with fast optimization algorithms to accelerate tensor contraction methods. A high-quality simulator that scales to sufficiently large circuits is one of the major bottlenecks in discovering applications for demonstrating quantum advantage.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

本项目的目标是开发一个专门的量子近似优化算法(QAOA)量子电路模拟器。QAOA是目前研究最多的量子优化算法，被认为是展示量子优势的首选算法。顶级量子信息科学研究人员正在进行一场世界性的竞赛，目的是寻找组合优化问题及其实例，这些问题和实例在量子设备上运行得更高效、更快，而不是在经典计算机上。关键的瓶颈之一是在经典计算机上更快地找到电路参数，以加速变分量子经典框架。对开发的模拟器的预期改进将使QAOA模拟的速度显著提高至少一个数量级，并显著加快寻找最佳QAOA电路参数的研究。因此，它将有助于美国科学家在这个竞争激烈的科学领域实现量子优势。该项目的技术目标是在开发可扩展的量子模拟器时，对张量压缩的各种节点消除方法进行计算和算法研究。该计划将建立在组合科学计算界在开发消除算法方面的成功的基础上，这些算法用于最小化矩阵计算中的错误和复杂性以及优化自动微分中的时间和空间复杂性，以及树宽优化算法的最新发展。此外，在这个项目中，重点将放在扩大相关的图形算法，以实现可接受的时间/质量折衷大规模量子模拟器。该项目的算法和软件产品将包括一个专门的QAOA量子电路开源模拟器，配备快速优化算法，以加速张量收缩方法。一个可扩展到足够大的电路的高质量模拟器是发现展示量子优势的应用程序的主要瓶颈之一。这一奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Constructing Optimal Contraction Trees for Tensor Network Quantum Circuit Simulation

• DOI: 10.1109/hpec55821.2022.9926353
• 发表时间: 2022-09
• 期刊: 2022 IEEE High Performance Extreme Computing Conference (HPEC)
• 作者: Cameron Ibrahim;Danylo Lykov;Zichang He;Y. Alexeev;Ilya Safro