CAREER: Enabling Scalable and Resilient Quantum Computer Architectures through Synergistic Hardware-Software Co-Design

量子计算机可以帮助解决物理、化学、材料设计、优化和机器学习中一些最复杂的问题。全世界都在努力创造更大、更可靠的量子计算机，提高它们执行复杂量子算法的能力。尽管取得了这一进展，但实际的量子计算仍面临着重大挑战，包括量子比特（qubit）设备对环境噪声的敏感性，从而产生不准确的结果。此外，可以在短期内利用量子计算机的算法的有限可用性是一个重大障碍，因为现有算法需要数百万次无噪声操作，超出了当前量子硬件的能力。解决现实世界问题所需的量子硬件与当今的量子计算技术之间的差距是巨大的。该奖项旨在通过硬件和软件设计的集成方法以及开发软件工具来帮助用户利用现有和未来的量子硬件，从而专注于可扩展和有弹性的量子计算机架构，从而弥合这一差距。此外，该奖项将创建一个广泛可访问的量子计算课程，并支持外展活动，以吸引和教育量子信息科学的本科生。该项目侧重于两个研究重点。第一个重点是共同设计指令集架构（ISA）和运行时，以实现高效和有弹性的架构，并引入新的硬件软件原语，以帮助扩展分布式量子位控制。在ISA方面，该项目将研究大多数量子硬件平台提供的灵活指令集如何在不过度增加校准复杂性的情况下利用，这些指令集通过校准脉冲来允许新的门。为了在容错量子计算机（FTQC）上实现大多数实际应用所需的大量操作，该项目将开发一个运行时，能够检测由于不稳定量子比特设备引起的噪声放大事件，并通过移动数据和学习最佳噪声缓解策略来减轻这些事件。该项目的第二个重点将集中在开发软件工具，以促进量子算法的有效经验设计和评估。为此，该项目将开发用于设计、调优和调试混合经典-量子工作流的工具。此外，该项目将研究部分容错量子计算机体系结构的有效性，其中只有一部分操作可以防止错误，并研究共同设计ISA、运行时、纠错和应用程序是否可以将部分FTQC推向实际用途。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Abstract Quantum computers can help solve some of the most complex problems in physics, chemistry, material design, optimization, and machine learning. A worldwide effort is underway to create larger, more reliable quantum computers, enhancing their ability to execute complex quantum algorithms. Despite this progress, practical quantum computing faces significant challenges, including the susceptibility of quantum bit (qubit) devices to environmental noise, producing inaccurate results. Furthermore, the limited availability of algorithms that can leverage quantum computers in the near term is a significant hurdle, as existing algorithms require millions of noise-free operations beyond the current quantum hardware capabilities. The gap between the quantum hardware necessary for solving real-world problems and today's quantum computing technology is significant. This award seeks to bridge this gap by focusing on scalable and resilient quantum computer architectures through an integrated approach to hardware and software design and developing software tools to help users leverage existing and future quantum hardware. Moreover, this award will create a broadly accessible quantum computing curriculum and support outreach activities to engage and educate undergraduates in quantum information sciences. This project focuses on two research thrusts. The first thrust concentrates on co-designing Instruction Set Architecture (ISA) and Runtime to enable efficient and resilient architectures and introduce new hardware-software primitives to help scale distributed qubit control. On the ISA front, the project will investigate how flexible instruction sets offered by most quantum hardware platforms, which allow new gates by calibrating pulses, can be leveraged without overly increasing the calibration complexity. For enabling a large number of operations on a Fault-Tolerant Quantum Computer (FTQC), required by most real-world applications, this project will develop a runtime capable of detecting noise amplification events due to unstable qubit devices and mitigating them by moving data and learning optimal noise mitigation policies. The second thrust of this project will focus on developing software tools to facilitate efficient empirical design and evaluation of quantum algorithms. To that end, the project will develop tools to design, tune, and debug hybrid classical-quantum workflows. Moreover, the project will study the efficacy of partially fault-tolerant quantum computer architectures, where only a subset of operations are protected against errors, and investigate if co-designing ISA, runtime, error correction, and applications can push partial FTQC toward practical utility.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.