XTRIPODS: Advancing Quantum Data Science Research and Education: Resilient Quantum Learning in NISQ era

XTRIPODS：推进量子数据科学研究和教育：NISQ 时代的弹性量子学习

• 批准号: 2343535
• 负责人: Ying Mao
• 金额: $ 20万
• 依托单位: Fordham University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2024
• 资助国家: 美国
• 起止时间: 2024-03-01 至 2026-02-28
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2343535&HistoricalAwards=false
• 关键词: XTRIPODS; Advancing; Quantum; Data; Science

Data science-based approaches have empowered popular technologies, including deep learning and large language models, leading to significant advancements in artificial intelligence. However, in the post-Moore’s Law era, the limitations of semiconductor fabrication, combined with the continuous growth of data sizes, have hindered further progress. Simultaneously, the rapid evolution of quantum computing has ushered in a new era, holding immense potential. This has sparked interest in quantum-based data science and learning, with the anticipation that certain systems could provide a quantum speedup. This project will harness the multi-disciplinary expertise from our NSF TRIPODS collaboration to develop noise-aware and resilient quantum learning systems. This project bridges the gap between research communities in traditional data science and quantum computing, thereby introducing classical data science researchers to the opportunities in quantum data science (QuanDS). Furthermore, this project will contribute to developing a skilled workforce, well-versed in cutting-edge AI and equipped to navigate the unique challenges of this emerging field in QuanDS.This project develops a quantum learning system using a data-driven approach that integrates static-dynamic combined circuit analysis. This method combines static metrics with the inherent dynamic noises of quantum systems by developing a set of noise-involved evaluation metrics, such as circuit sensitivity indicator. A topology-aware resilient circuit transpilation mechanism tailored to quantum learning will then be developed to optimize the execution on noisy intermediate-scale quantum (NISQ) era hardware under quantum noise. Additionally, the project provides a holistic analysis of the vulnerability in distributed quantum learning by reasoning the strategic behavior of the attacker to quantum computing nodes, and devises resilient learning countermeasures in the distributed quantum-classical learning paradigm. Furthermore, it develops educational modules in quantum control systems combined with resilient quantum learning techniques. The comprehensive outreach activities will be developed to integrate QuanDS research and education.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.

基于数据科学的方法为包括深度学习和大型语言模型在内的流行技术提供了支持，从而使人工智能取得了重大进展。然而，在后摩尔定律时代，半导体制造的局限性，加上数据规模的不断增长，阻碍了进一步的发展。与此同时，量子计算的快速发展开创了一个新时代，拥有巨大的潜力。这引发了人们对基于量子的数据科学和学习的兴趣，人们期待某些系统可以提供量子加速。该项目将利用我们与NSF TRIPODS合作的多学科专业知识来开发噪声感知和弹性量子学习系统。该项目弥合了传统数据科学和量子计算研究社区之间的差距，从而将经典数据科学研究人员引入量子数据科学（QuanDS）的机会。此外，该项目将有助于培养熟练的劳动力，精通尖端人工智能，并有能力在QuanDS中应对这一新兴领域的独特挑战。该项目使用数据驱动的方法开发量子学习系统，该方法集成了静态-动态组合电路分析。该方法将量子系统固有的动态噪声与静态噪声相结合，提出了一套包含噪声的评价指标，如电路灵敏度指标。然后，将开发一种针对量子学习的拓扑感知弹性电路转译机制，以优化量子噪声下噪声中间尺度量子（NISQ）时代硬件的执行。此外，该项目通过推理攻击者对量子计算节点的策略行为，对分布式量子学习中的漏洞进行了全面分析，并在分布式量子经典学习范式中设计了弹性学习对策。此外，它还开发了量子控制系统中的教育模块，并结合了弹性量子学习技术。该奖项反映了NSF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。