CAREER: Photonic Quantum Machine Learning: From Architecture to Applications

职业：光子量子机器学习：从架构到应用

• 批准号: 2144057
• 负责人: Zheshen Zhang
• 金额: $ 50万
• 依托单位: University of Arizona
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-03-15 至 2023-03-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2144057&HistoricalAwards=false
• 关键词: CAREER; Photonic; Quantum; Machine; Learning

The formulation of quantum mechanics in the 20th century shifted the landscape of science, giving birth to a plethora of revolutionary technologies that empower the information age. Humankind is now on the verge of a second quantum revolution fueled by quantum information science (QIS), which is envisioned to enable disruptive communication, sensing, and computing applications. Despite the tremendous prospects promised by QIS, building large-scale and robust quantum information processing systems remains an outstanding challenge due to the fragility of quantum information in the present noisy intermediate-scale quantum (NISQ) hardware. To unlock the power of NISQ devices and systems, hybrid quantum-classical protocols have become a focus of recent QIS studies, in which state-of-the-art classical data science tools are leveraged to steer NISQ hardware towards solving specific data-processing problems. This CAREER project will develop a new photonic quantum machine-learning architecture that combines mature, classical machine-learning tools and NISQ platforms to endow unprecedented communication, sensing, and data processing capabilities. Compared with other NISQ platforms, quantum photonics feature room-temperature operations, mass productivity, and compatibility with the existing telecommunication and sensing infrastructures. The project will advance basic knowledge for the NISQ era and the interdisciplinary areas of QIS, machine learning, and NSF’s 10 Big Ideas Harnessing the Data Revolution and the Quantum Leap. A critical ingredient for a sustainable QIS ecosystem is to develop the next-generation quantum workforce. To this end, this CAREER project will encompass activities for: 1) QIS teaching laboratories for undergraduate students; 2) a QIS training program for industry workforce development; and 3) outreach to engage K-12 STEM students early in QIS.The research activities of this CAREER project will encompass both: 1) a photonic quantum machine-learning architecture based on a classical machine-learning framework and photonic quantum information-processing hardware, including reconfigurable entanglement sources and adaptive quantum receivers; and 2) photonic quantum machine-learning applications for long-haul optical communications, multi-domain sensing, and quantum-enhanced data processing. The new photonic quantum machine-learning architecture will effectively use cutting-edge classical machine-learning tools to configure variational photonic quantum circuits, as a powerful means to generate, process, and measure quantum information. Although photons interact only weakly with each other to hinder the use of large-scale photonic entanglement, the proposed photonic quantum machine-learning architecture will overcome this barrier by leveraging quantum photonics that offer deterministic generation, the processing of large-scale entanglement, and suitability for sensing and communication applications. By combining enhancements from machine learning and quantum coherence, the expected project outcomes will enhance various sensing- and communication-related tasks, including pattern recognition, deep-space signal detection, and efficient data compression. Ultimately, broadly sharing the new knowledge grown out of these research activities should spark collaborations between academia, National Laboratories, and the U.S. healthcare, aerospace, environmental protection, and chemical engineering industries. By working with industrial partners, the CAREER project will connect with new quantum technologies that transform U.S. industries. The project research facilities and workforce development activities will help to prepare the U.S. technology industry workforce for the future quantum edge and establish comfortable and personally relevant quantum foundations for students across university to high-school settings.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.

量子力学在20世纪的形成改变了科学的面貌，催生了大量的革命性技术，赋予信息时代以力量。人类现在正处于由量子信息科学（QIS）推动的第二次量子革命的边缘，量子信息科学被设想为能够实现破坏性的通信，传感和计算应用。尽管量子信息系统有着巨大的发展前景，但由于量子信息在现有的噪声中尺度量子（NISQ）硬件中的脆弱性，构建大规模和鲁棒的量子信息处理系统仍然是一个突出的挑战。为了释放NISQ设备和系统的力量，混合量子经典协议已成为最近QIS研究的焦点，其中利用最先进的经典数据科学工具来引导NISQ硬件解决特定的数据处理问题。这个CAREER项目将开发一种新的光子量子机器学习架构，结合成熟的经典机器学习工具和NISQ平台，赋予前所未有的通信、传感和数据处理能力。与其他NISQ平台相比，量子光子学具有室温操作，大规模生产力以及与现有电信和传感基础设施的兼容性。该项目将推进NISQ时代的基础知识，以及QIS，机器学习和NSF利用数据革命和量子飞跃的10大想法的跨学科领域。可持续QIS生态系统的一个关键因素是开发下一代量子劳动力。为此，该CAREER项目将包括以下活动：1）本科生的QIS教学实验室; 2）行业劳动力发展的QIS培训计划;以及3）在QIS早期接触K-12 STEM学生的外联活动。该CAREER项目的研究活动将包括：1）基于经典机器学习框架和光子量子信息处理硬件的光子量子机器学习架构，包括可重新配置的纠缠源和自适应量子接收器;以及2）用于长距离光通信、多域感测和量子增强数据处理的光子量子机器学习应用。新的光子量子机器学习架构将有效地使用尖端的经典机器学习工具来配置变化的光子量子电路，作为生成，处理和测量量子信息的强大手段。虽然光子之间的相互作用很弱，阻碍了大规模光子纠缠的使用，但所提出的光子量子机器学习架构将通过利用量子光子学来克服这一障碍，量子光子学提供确定性生成，大规模纠缠的处理以及传感和通信应用的适用性。通过结合机器学习和量子相干的增强，预期的项目成果将增强各种传感和通信相关的任务，包括模式识别，深空信号检测和有效的数据压缩。最终，广泛分享这些研究活动产生的新知识应该会激发学术界、国家实验室和美国医疗保健、航空航天、环境保护和化学工程行业之间的合作。通过与工业合作伙伴的合作，CAREER项目将与改变美国工业的新量子技术联系起来。该项目的研究设施和劳动力发展活动将有助于美国技术行业的劳动力为未来的量子优势做好准备，并为大学到高中的学生建立舒适和个人相关的量子基础。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。