REU Site: Secure, Robust, and Resilient AI-enabled System Engineering

REU 站点：安全、稳健且有弹性的人工智能系统工程

• 批准号: 2050972
• 负责人: Ming Shao
• 金额: $ 40.46万
• 依托单位: University of Massachusetts, Dartmouth
• 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=2050972&HistoricalAwards=false
• 关键词: REU; Site; Secure; Robust; Resilient

Artificial intelligence (AI) has gained significant attention in recent years as an enabler of technologies such as autonomous cyber-physical systems and cybersecurity. This funding establishes a new Research Experiences for Undergraduates (REU) Site at the University of Massachusetts Dartmouth with a focus on secure, robust, and resilient AI-enabled system engineering. In each summer, ten undergraduate students will participate in research for ten weeks. During the research, participants will learn engineering principles to support the rigorous design and quantitative test of systems and incorporate AI components to ensure properties such as security, robustness, and resilience. This program will provide a research experience with a balance of theory, hands-on practical skills, and applications. The project is led by a team of experienced faculty who will mentor the participating researchers through a number of professional development activities, which prepare them to take a systems perspective while maintaining technical excellence to enhance the robustness of AI-enabled systems in commercial and societal contexts. This project will provide unique opportunities for undergraduate students, especially for those from underrepresented and minority groups and institutions with limited research resources, to participate in real-world research.Artificial intelligence has gained significant attention in recent years as an enabler of technologies such as autonomous cyber physical systems and cyber security. However, engineering principles to support the rigorous design and quantitative test of systems incorporating AI components are needed to ensure properties such as security, robustness, and resilience. AI system engineering research will provide the foundation required to establish user trust and promote regulatory activities such as safety assurance. To enable to the promise of AI systems, this project proposes a unique Research Experiences for Undergraduates (REU) Site focused on secure, robust, and resilient AI-enabled system engineering. This REU Site will intertwine AI research with cybersecurity and systems engineering in a portfolio of synergistic projects. Ten undergraduate students will be engaged in a ten-week summer research program to conduct research at the University of Massachusetts Dartmouth (UMD). The themes of the proposed REU Site are AI, cybersecurity, and system engineering research and their combined implications. Specifically, the following research topics will be explored: (1) cyber attacks that degrade the performance of AI algorithms; (2) AI-enabled cybersecurity against malicious actors; (3) resilient AI-enabled system engineering methods. These research themes will create opportunities for multidisciplinary knowledge sharing between AI and cybersecurity from software, hardware and system engineering perspectives, including topics in biometrics, UAVs, brain computer interaction, and signal processing. The REU will prepare talented young researchers to take a systems perspective, while maintaining technical excellence to enhance the robustness of AI-enabled systems in commercial and societal contexts. The site will provide unique opportunities for undergraduate students, especially for those from underrepresented and minority groups and institutions with limited research resources, to participate in real-world research.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.

近年来，人工智能（AI）作为自主网络物理系统和网络安全等技术的推动者受到了广泛关注。这笔资金在马萨诸塞州大学达特茅斯建立了一个新的本科生研究经验（REU）网站，重点关注安全，强大和有弹性的人工智能系统工程。每年夏天，十名本科生将参加为期十周的研究。在研究过程中，参与者将学习工程原理，以支持系统的严格设计和定量测试，并纳入人工智能组件，以确保安全性，鲁棒性和弹性等属性。该计划将提供理论，实践技能和应用平衡的研究经验。该项目由经验丰富的教师团队领导，他们将通过一些专业发展活动指导参与研究人员，使他们能够从系统的角度出发，同时保持技术卓越，以增强商业和社会环境中人工智能系统的稳健性。该项目将为本科生提供独特的机会，特别是那些来自代表性不足和少数群体以及研究资源有限的机构的学生，参与现实世界的研究。人工智能近年来作为自主网络物理系统和网络安全等技术的推动者而受到广泛关注。然而，需要支持包含AI组件的系统的严格设计和定量测试的工程原则，以确保安全性，鲁棒性和弹性等属性。人工智能系统工程研究将为建立用户信任和促进安全保证等监管活动提供必要的基础。为了实现人工智能系统的承诺，该项目提出了一个独特的本科生研究经验（REU）网站，专注于安全，强大和有弹性的人工智能系统工程。这个REU网站将在一系列协同项目中将AI研究与网络安全和系统工程相结合。10名本科生将参加为期10周的夏季研究计划，在马萨诸塞州达特茅斯大学（UMD）进行研究。 拟议的REU网站的主题是人工智能，网络安全和系统工程研究及其综合影响。具体而言，将探讨以下研究主题：（1）降低人工智能算法性能的网络攻击;（2）针对恶意行为者的人工智能网络安全;（3）弹性人工智能系统工程方法。这些研究主题将从软件，硬件和系统工程的角度为人工智能和网络安全之间的多学科知识共享创造机会，包括生物识别，无人机，脑机交互和信号处理等主题。REU将培养有才华的年轻研究人员从系统的角度出发，同时保持技术卓越，以增强商业和社会环境中人工智能系统的鲁棒性。该网站将为本科生提供独特的机会，特别是那些来自代表性不足和少数群体以及研究资源有限的机构的学生，参与现实世界的研究。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。

项目成果

A Novel Convolutional Neural Network for Emotion Recognition Using Neurophysiological Signals

• DOI: 10.1109/icra46639.2022.9811868
• 发表时间: 2022-05
• 期刊: 2022 International Conference on Robotics and Automation (ICRA)
• 作者: Marc Tunnell;Huijin Chung;Yuchou Chang

Sensor Data-Driven UAV Anomaly Detection using Deep Learning Approach

• DOI: 10.1109/milcom52596.2021.9653036
• 发表时间: 2021-11
• 期刊: MILCOM 2021 - 2021 IEEE Military Communications Conference (MILCOM)
• 作者: Julio Galvan;A. Raja;Yanyan Li;Jiawei Yuan

Machine-Learning PUF-based Detection of RF Anomalies in a Cluttered RF Environment

杂乱 RF 环境中基于机器学习 PUF 的 RF 异常检测

• DOI: 10.1109/hst53381.2021.9619834
• 发表时间: 2021
• 期刊: 2021 IEEE International Symposium on Technologies for Homeland Security (HST
• 作者: Lu, James;Morehouse, Todd;Yuan, Jiawei;Zhou, Ruolin
• 通讯作者: Zhou, Ruolin

Adversary on Multimodal BCI-based Classification

基于多模式 BCI 分类的对手

• 发表时间: 2023
• 期刊: 11th International IEEE EMBS Conference on Neural Engineering
• 作者: Kumar, Chetan;Donohue, James P.;Gonjari, Rohan;Rahimi, Neela;McLinden, John;Shahriari, Yalda;Shao, Ming
• 通讯作者: Shao, Ming

Visual Inference Using Homology of Human and Machine Vision Systems

利用人类和机器视觉系统的同源性进行视觉推理

• 发表时间: 2022
• 期刊: Advances in cognitive systems
• 作者: Morosco, Jacob W.;Chang, Yuchou
• 通讯作者: Chang, Yuchou