Collaborative Research: EPiQC: Enabling Practical-Scale Quantum Computation

合作研究：EPiQC：实现实用规模的量子计算

• 批准号: 1729369
• 负责人: Peter Shor
• 金额: $ 270万
• 依托单位: Massachusetts Institute of Technology
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-03-01 至 2025-02-28
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1729369&HistoricalAwards=false
• 关键词: Collaborative; Research; EPiQC; Enabling; Practical

Quantum computing sits poised at the verge of a revolution. Quantum machines may soon be capable of performing calculations in machine learning, computer security, chemistry, and other fields that are extremely difficult or even impossible for today's computers. Few of these limitless possibilities on the horizon that quantum computing could lead to are better drug discovery, more efficient photovoltaics, new nanoscale materials, and perhaps even more efficient food production. These benefits will be enabled by substantially improving the ability to solve computational problems in quantum chemistry, quantum simulation, and optimization. These dramatic improvements arise because each additional quantum bit doubles the potential computing power of a machine, accumulating exponential gains that could eventually eclipse the world's largest supercomputers. Quantum computing will also drive a new segment of the computing industry, providing new strategies for specific applications that increase computational power even as physical limits slow improvements in classical silicon-chip technology. This multi-institutional project, Enabling Practical-scale Quantum Computing (EPiQC) Expedition, will help bring the great potential of this new paradigm into reality by reducing the current gap between existing theoretical algorithms and practical quantum computing architectures. Over five years, the EPiQC Expedition will collectively develop new algorithms, software, and machine designs tailored to key properties of quantum device technologies with 100 to 1000 quantum bits. This work will facilitate profound new scientific discoveries and also broadly impact the state of high-performance computing. To prepare the U.S. workforce for this revolution in computing, we need to educate citizens to think about computing from a quantum perspective, integrating concepts such as probability and uncertainty into the digital lexicon. The EPiQC Expedition will design teaching curricula and distribute exemplar materials for students ranging from primary school to engineers in industry. EPiQC will also establish an academic-industry consortium which will share educational and research products and accelerate the pace of quantum computing design and applications. Because quantum computing is a new branch of computer science, it will require entirely new types of algorithms and software. In order to produce practical quantum computation in the near future, these elements cannot be developed in isolation. Instead, researchers must increase the efficiency of quantum algorithms running on quantum machines through the simultaneous design and optimization of algorithms, software and machines. New algorithms and software need to know what specific machine operations are easy or difficult in a given quantum technology and must be prepared to produce useful answers from imperfect results from imperfect machines. Software also needs to verify that the computation executed correctly as expected, an especially difficult task given that conventional machines cannot simulate even a modest-size quantum machine. The EPiQC Expedition unites experts on algorithms, software, architecture, and education to develop these elements in parallel. Overall, EPiQC will increase the efficiency of practical quantum computations by 100 to 1000 times, effectively bringing quantum computing out of the laboratory and into practical use 10-20 years sooner than through technology advances alone. The project identifies 4 thrusts: algorithmic innovations, compiler development, verification, and the broader impact tasks of developing education modules. The algorithmic tasks are organized into the subdomains of optimization, computational chemistry, and the discovery of separations between quantum and classical speedup. The compiler tasks are more milestone driven - development of technology libraries, development of various compilation techniques which leverages these libraries, as well as novel error correction schemes. The project will tie the tool chain closely to the underlying hardware and fault-tolerance mechanisms.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.

量子计算正处于一场革命的边缘。量子机器可能很快就能在机器学习、计算机安全、化学和其他对今天的计算机来说极其困难甚至不可能的领域进行计算。量子计算可能带来的这些无限可能性中，很少有更好的药物发现、更高效的光伏、新的纳米级材料，甚至可能更有效的食品生产。这些好处将通过大幅提高解决量子化学、量子模拟和优化中的计算问题的能力来实现。这些戏剧性的改进之所以出现，是因为每增加一个量子比特，机器的潜在计算能力就会翻倍，积累的指数级增长最终可能会让世界上最大的超级计算机黯然失色。量子计算还将推动计算行业的一个新领域，为提高计算能力的特定应用提供新的策略，即使传统硅芯片技术的物理限制减缓了改进。这个多机构项目，实现实用规模量子计算（EPiQC）远征，将通过缩小现有理论算法和实际量子计算架构之间的当前差距，帮助将这种新范式的巨大潜力变为现实。在五年内，EPiQC远征将共同开发新的算法、软件和机器设计，以适应100到1000量子比特的量子器件技术的关键特性。这项工作将促进深刻的新科学发现，并广泛影响高性能计算的状态。为了让美国劳动力为这场计算革命做好准备，我们需要教育公民从量子的角度思考计算，将概率和不确定性等概念整合到数字词典中。EPiQC考察将设计教学课程，并为从小学到工业工程师的学生分发范例材料。EPiQC还将建立一个学术-产业联盟，共享教育和研究产品，加快量子计算设计和应用的步伐。由于量子计算是计算机科学的一个新分支，它将需要全新类型的算法和软件。为了在不久的将来产生实用的量子计算，这些元素不能孤立地发展。相反，研究人员必须通过同时设计和优化算法、软件和机器来提高在量子机器上运行的量子算法的效率。新的算法和软件需要知道在给定的量子技术中哪些特定的机器操作是容易的或困难的，并且必须准备好从不完美的机器的不完美结果中产生有用的答案。软件还需要验证计算是否按预期正确执行，这是一项特别困难的任务，因为传统机器甚至无法模拟中等大小的量子机器。EPiQC远征联合了算法、软件、架构和教育方面的专家，并行开发这些元素。总体而言，EPiQC将使实际量子计算的效率提高100到1000倍，有效地使量子计算走出实验室并投入实际使用，比仅通过技术进步提前10-20年。该项目确定了4个重点：算法创新、编译器开发、验证以及开发教育模块的更广泛影响任务。算法任务被组织成优化、计算化学和发现量子加速与经典加速之间的分离的子领域。编译器任务更多的是里程碑驱动的——技术库的开发，利用这些库的各种编译技术的开发，以及新的纠错方案。该项目将把工具链与底层硬件和容错机制紧密联系起来。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。

项目成果

Improved graph formalism for quantum circuit simulation

改进量子电路模拟的图形形式

• DOI: 10.1103/physreva.105.022432
• 发表时间: 2022
• 期刊: Physical Review A
• 影响因子: 2.9
• 作者: Hu, Alexander Tianlin;Khesin, Andrey Boris
• 通讯作者: Khesin, Andrey Boris

Exploring ququart computation on a transmon using optimal control

• DOI: 10.1103/physreva.108.062609
• 发表时间: 2023-04
• 期刊: Physical Review A
• 影响因子: 2.9
• 作者: Lennart Maximilian Seifert;Ziqian Li;Tanay Roy;D. Schuster;F. Chong;Jonathan M. Baker

QuantumNAS: Noise-Adaptive Search for Robust Quantum Circuits

• DOI: 10.1109/hpca53966.2022.00057
• 发表时间: 2021-07
• 期刊: 2022 IEEE International Symposium on High-Performance Computer Architecture (HPCA)
• 作者: Hanrui Wang;Yongshan Ding;Jiaqi Gu;Yujun Lin;D. Pan;F. Chong;Song Han

Coreset Clustering on Small Quantum Computers

• DOI: 10.3390/electronics10141690
• 发表时间: 2020-04
• 期刊: ArXiv
• 作者: T. Tomesh;P. Gokhale;Eric R. Anschuetz;F. Chong

Classical algorithms, correlation decay, and complex zeros of partition functions of quantum many-body systems

量子多体系统配分函数的经典算法、相关衰减和复零点

• DOI: 10.1145/3357713.3384322
• 发表时间: 2020
• 期刊: STOC 2020: Proceedings of the 52nd Annual ACM SIGACT Symposium on Theory of Computing
• 作者: Harrow, Aram W.;Mehraban, Saeed;Soleimanifar, Mehdi
• 通讯作者: Soleimanifar, Mehdi