REU Site: Online Interdisciplinary Big Data Analytics in Science and Engineering

REU 网站：科学与工程领域的在线跨学科大数据分析

• 批准号: 2050943
• 负责人: Jianwu Wang
• 金额: $ 29.37万
• 依托单位: University of Maryland Baltimore County
• 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: 2021-04-01 至 2024-03-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2050943&HistoricalAwards=false
• 关键词: REU; Site; Online; Interdisciplinary; Big

This REU Site program will provide 8-week summer online research experiences to undergraduates on how to utilize modern data science and high-performance computing (HPC) techniques to process and analyze big data in many science and engineering disciplines such as Atmospheric Science, Mechanical Engineering, and Medicine. The REU Site program will be conducted purely online to allow students to conduct research without traveling and to work with experts nationwide. In recent years, the astronomical growth of available datasets in many science and engineering disciplines often requires big data analytics techniques to efficiently and effectively process the large datasets and obtain knowledge and insight from them. The program will help students identify frontier research challenges encountered when facing big data in science and engineering, and guide students to conduct research to tackle those challenges using advanced cyberinfrastructure software technologies (big data, distributed machine/deep learning, HPC, etc.) and hardware resources (including big data clusters, CPU clusters and GPU clusters). The program will develop the national workforce in areas of critical need on “Data + Computing + X”. The project thus serves the national interest, as stated by NSF's mission, to promote the progress of science and advance the national prosperity and welfare. By having three phases of training, namely formal instruction, team-based research and dissemination, this REU Site program will i) ignite students' interest in how data science and high-performance computing techniques can aid in the scientific discovery process via interdisciplinary research projects; ii) provide interdisciplinary team-based research experiences via guidance from research mentors, graduate assistants, and interdisciplinary collaborators in the application domain area of each project; and iii) provide integrated training on crucial professional skills (including skills in collaboration, communication, presentation and writing, and experience in scientific paper preparation and presentation). This REU Site will provide a unique and comprehensive program integrating the following elements under the guidance of the faculty team, who have extensive experience in interdisciplinary research and online education: 1) frontier research connecting advanced cyberinfrastructure techniques with big data challenges in science and engineering, 2) online research experience that leverages modern communication tools, 3) team-based interdisciplinary research and communication experience, 4) complementary professional development activities (such as graduate school preparation, invited talks from established researchers, and interaction with university administers including college deans and department chairs), 5) educational research on online undergraduate training experiences.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.

REU网站项目将为本科生提供为期8周的暑期在线研究体验，学习如何利用现代数据科学和高性能计算（HPC）技术处理和分析许多科学和工程学科（如大气科学、机械工程和医学）中的大数据。REU网站计划将完全在线进行，让学生无需旅行就可以进行研究，并与全国专家一起工作。近年来，许多科学和工程学科中可用数据集的天文数字增长往往需要大数据分析技术来高效和有效地处理大数据集，并从中获得知识和见解。该计划将帮助学生识别在科学和工程中面对大数据时遇到的前沿研究挑战，并指导学生利用先进的网络基础设施软件技术（大数据，分布式机器/深度学习，HPC等）和硬件资源（包括大数据集群，CPU集群和GPU集群）进行研究以解决这些挑战。该计划将在“数据+计算+ X”的关键需求领域培养国家劳动力。因此，该项目符合国家利益，正如美国国家科学基金会的使命所述，促进科学进步，促进国家繁荣和福利。通过三个阶段的培训，即正式指导，团队研究和传播，REU网站计划将i)激发学生对数据科学和高性能计算技术如何通过跨学科研究项目帮助科学发现过程的兴趣；Ii)通过每个项目应用领域的研究导师、研究生助理和跨学科合作者的指导，提供跨学科团队研究经验；提供关键专业技能的综合培训（包括协作、沟通、演示和写作技能，以及科学论文准备和演示的经验）。这个REU网站将提供一个独特而全面的项目，在教师团队的指导下，整合以下要素，他们在跨学科研究和在线教育方面拥有丰富的经验：1)将先进的网络基础设施技术与科学和工程领域的大数据挑战相结合的前沿研究；2)利用现代通信工具的在线研究经验；3)基于团队的跨学科研究和交流经验；4)互补的专业发展活动（如研究生院准备，知名研究人员的邀请演讲，以及与包括学院院长和系主任在内的大学管理人员的互动）；5)本科在线培训体验教育研究。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

Sea Ice Forecasting using Attention-based Ensemble LSTM

• DOI: 10.13016/m2gbqc-q97l
• 发表时间: 2021-07
• 期刊: ArXiv
• 作者: Sahara Ali;Yiyi Huang;Xin Huang;Jianwu Wang

Atmospheric Gravity Wave Detection Using Transfer Learning Techniques

• DOI: 10.1109/bdcat56447.2022.00023
• 发表时间: 2022-12
• 期刊: 2022 IEEE/ACM International Conference on Big Data Computing, Applications and Technologies (BDCAT)
• 作者: Jorge L. Gonzalez;Theodore Chapman;Kathryn Chen;Hannah M. Nguyen;Logan Chambers;S. A. Mostafa

Promising Hyperparameter Configurations for Deep Fully Connected Neural Networks to Improve Image Reconstruction in Proton Radiotherapy

有前景的深度全连接神经网络超参数配置可改善质子放射治疗中的图像重建

• 发表时间: 2021
• 期刊: and Security (REU 2021 Symposium
• 作者: York, Sokhna;Ali, Alina;Lashbrooke, David;Yepez-Lopez, Rodrigo;Barajas, Carlos;Gobbert, Matthias;Polf, Jerimy
• 通讯作者: Polf, Jerimy

Deep residual fully connected neural network classification of Compton camera based prompt gamma imaging for proton radiotherapy

• DOI: 10.3389/fphy.2023.903929
• 发表时间: 2023-02-16
• 期刊: FRONTIERS IN PHYSICS
• 影响因子: 3.1
• 作者: Barajas, Carlos A. A.;Polf, Jerimy C. C.;Gobbert, Matthias K. K.
• 通讯作者: Gobbert, Matthias K. K.

Multi-Layer Recurrent Neural Networks for the Classification of Compton Camera Based Imaging Data for Proton Beam Cancer Treatment

用于质子束癌症治疗的基于康普顿相机的成像数据分类的多层循环神经网络

• 发表时间: 2023
• 期刊: Applications and Technologies (BDCAT 2022
• 作者: Clark, Joseph;Gaillard, Anaise;Koe, Justin;Navarathna, Nithya;Kelly, Daniel;Gobbert, Matthias;Barajas, Carlos;Polf, Jerimy
• 通讯作者: Polf, Jerimy