Collaborative Research: PIF: Scalable Quantum Computers in the Presence of Physical Noise: a Study of Surface Codes with Realistic Errors at the Algorithmic Level

原则上，量子计算机可以使用原子和分子的物理来执行从密码破译到材料模拟的各种计算，在算法上比任何传统的超级计算机都要快。建造量子计算机的主要挑战是量子组件容易出错。纠错可以用来克服这一挑战，但它对未来的量子计算机硬件提出了严格的要求。这项研究的主要目标是为弥合当前物理硬件和大规模量子算法之间的差距开辟一条道路。具体地说，我们将研究一种最先进的量子纠错方案，即表面代码，在映射到现实的物理架构时是如何表现的。如果成功，这项研究将帮助实验者建造第一台通用的可扩展量子计算机。这项研究将培训不断增长的量子信息科学领域的研究生和本科生，并为他们提供与IBM的行业合作者合作的机会。这一提议通过探索实验设备、计算机体系结构和量子信息理论的交集来探索量子计算，从而扩展了物理学的前沿。计算误差阈值为1%的表面编码是解决量子系统中退相干和缺陷问题的一种很有前途的方案。计算是通过创建、湮灭和编织拓扑缺陷来执行的。从算法和体系结构的角度研究这一过程将有助于揭示共同的结构，并为在真实的物理系统上实现这些想法提供一种精确的方法。我们将研究考虑真实噪声模型和系统级约束的表面代码实现。这将利用之前在实验设备数据、量子电路综合工具、计算机体系结构以及量子编程语言和编译器方面的工作。此外，还将与IBM的Sergey Bravyi就这项工作的理论基础进行密切合作(包括在IBM为学生提供暑期实习)。

In principle, quantum computers can use the physics of atoms and molecules to perform calculations ranging from code-breaking to simulation of materials algorithmically faster than any conventional supercomputer. The main challenge for building a quantum computer is that quantum components are prone to error. Error correction can be used to overcome this challenge but it places stringent requirements on future quantum computer hardware. The primary goal of this research is to develop a path for bridging the gap between current physical hardware and large-scale quantum algorithms. Specifically how a state-of-the-art quantum error correction scheme, the surface code, behaves when mapped to realistic physical architectures will be examined. If successful, this research will help experimentalists build the first universal scalable quantum computer. The research will train graduate students and undergraduate students in the growing field of quantum information science and provide them with an opportunity to work with industrial collaborators at IBM. This proposal expands the frontiers of physics by exploring quantum computation at the intersection of experimental devices, computer architecture, and quantum information theory. Surface codes with a computational error threshold of 1% are a promising solution to the problem of decoherence and imperfections in quantum systems. Computation is performed by the creation, annihilation, and braiding of topological defects. Examining this process from the perspective of algorithms and architectures will help reveal common structures and provide a precise method for implementation of these ideas on real physical systems. Surface code implementations that account for realistic noise models and system-level constraints will be examined. This will leverage previous work in experimental device data, quantum circuit synthesis tools, computer architecture, and quantum programming languages and compilers. Additionally, there will be close collaboration with Sergey Bravyi at IBM (including summer internships at IBM for students) on the theoretical foundations of this work.