REU Site: Interdisciplinary Research Experience on Accelerated Deep Learning through A Hardware-Software Collaborative Approach

REU 网站：通过硬件-软件协作方法加速深度学习的跨学科研究经验

• 批准号: 2051062
• 负责人: Hui Zhao
• 金额: $ 39.87万
• 依托单位: University of North Texas
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-03-01 至 2024-02-29
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2051062&HistoricalAwards=false
• 关键词: REU; Site; Interdisciplinary; Research; Experience

This Research Experiences for Undergraduates (REU) Site Program at the University of North Texas will enhance the knowledge and research skills of a diverse cohort of undergraduate students through empowering, innovative, and interdisciplinary research experiences in developing Deep Learning applications and systems. The program aims to 1) expose undergraduate students to real-world and cutting-edge research focused on accelerated deep learning through combined hardware and software development; 2) encourage more undergraduate students to continue their academic careers and seek graduate degrees in computer science, computer engineering, and related disciplines; 3) develop research skills and improve communication and collaborative skills in undergraduate students. The emphasis of this program is on the interdisciplinary research experience for the REU students to gain hardware-software codesign skills. REU students will be mentored by experienced researchers in the area of AI/machine learning algorithms, computer system and architecture, biomedical informatics and computational life science. Ten students will be hosted on the University of North Texas campus each summer where they closely collaborate with faculty and graduate students in performing innovative research that will address the timely and critical problems in developing Deep Learning systems and algorithms. Through their projects, the students will be able to observe the connections and interactions between different disciplines involving hardware and software. The REU students will also be formally trained in the process of collaborative and responsible research including ethics, teamwork and leadership. The program will also help the students to improve their communication abilities and writing skills through activities such as oral presentations and technical reports. Through this program, the students will be empowered with research skills and experience to empower their journey to become successful researchers.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.

北德克萨斯大学的本科生研究经验（REU）现场项目将通过在开发深度学习应用和系统方面的授权、创新和跨学科研究经验，提高不同本科生群体的知识和研究技能。该项目旨在1)通过结合硬件和软件开发，让本科生接触到专注于加速深度学习的现实世界和前沿研究；2)鼓励更多本科生继续其学术生涯，攻读计算机科学、计算机工程及相关学科的研究生学位；3)培养大学生的研究能力，提高学生的沟通和协作能力。本课程的重点是为REU学生提供跨学科的研究经验，以获得硬件软件协同设计技能。REU的学生将受到人工智能/机器学习算法、计算机系统和架构、生物医学信息学和计算生命科学领域经验丰富的研究人员的指导。每年夏天，将有10名学生在北德克萨斯大学的校园里与教师和研究生密切合作，进行创新研究，解决开发深度学习系统和算法中的及时和关键问题。通过他们的项目，学生将能够观察到涉及硬件和软件的不同学科之间的联系和互动。REU的学生还将在合作和负责任的研究过程中接受正式培训，包括道德、团队合作和领导能力。该项目还将通过口头报告和技术报告等活动，帮助学生提高沟通能力和写作技巧。通过这个项目，学生们将获得研究技能和经验，使他们成为成功的研究人员。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

Genomics-GPU: A Benchmark Suite for GPU-accelerated Genome Analysis

• DOI: 10.1109/ispass57527.2023.00026
• 发表时间: 2023-04
• 期刊: 2023 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)
• 作者: Zhuren Liu;Shouzhe Zhang;Justin Garrigus;Hui Zhao