REU Site: Data-driven Security

该奖项建立了一个新的本科生研究经验（REU）网站，专注于博伊西州立大学的数据驱动安全。数据驱动安全是一个新兴的跨学科领域，它应用数据科学和人工智能来减轻网络攻击和其他安全风险和威胁。本科生将与来自计算机科学和数学学科的教师导师一起参加暑期研究活动。学生们将以团队形式探索重要的研究问题，并将参加其他专业发展活动，为他们未来在计算机领域的职业生涯做好准备。该网站将面向传统上在计算机科学领域代表性不足的群体的学生，以及来自太平洋西北地区两年制大学的学生。REU网站将以数据驱动安全领域的变革性跨学科研究为特色。研究问题包括：一种新的数据驱动方法来学习联盟博弈的特征函数，该方法模拟了一个隐蔽网络，以提高关键行为者的识别；发现不当行为和减少错误信息的新方法；基于鲁棒性的企业网络异常检测流量行为模型提取开发新的轻量级加密算法，对算法弱点和通过机器学习执行的侧信道攻击都具有鲁棒性。所有的研究都需要数据科学、人工智能、数学和安全的深度融合。学生将学习在团队中工作，并通过参与基于建设性对话和协作设计的调查方法的活动，将他们的结果传达给不同的受众。这些项目有可能扩大几个领域的知识，包括博弈论、入侵检测系统、错误信息缓解和轻量级密码学。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。

This award establishes a new Research Experiences for Undergraduates (REU) Site focused on data-driven security at Boise State University. Data-driven security is an emerging interdisciplinary field that applies data science and artificial intelligence to mitigate cyberattacks and other security risks and threats. Undergraduate students will participate in summer research activities with faculty mentors from the computer science and mathematics disciplines. The students will work in teams to explore important research questions and will also participate in other professional development activities that will prepare them for future careers in the computing fields. The site will target students from groups traditionally under-represented in computer science as well as students from two-year colleges in the Pacific Northwest.The REU Site will feature transformative interdisciplinary research in the field of data-driven security. Research problems include: a novel data-driven approach to learn the characteristic function of a coalition game that models a covert network to improve key actor identification; new approaches to detect misbehavior and mitigate misinformation; innovative robust traffic-flow behavioral model extraction for detecting anomalous situations in enterprise networks; development of new lightweight cryptographic algorithms that are robust to both algorithmic weaknesses and side-channel attacks performed with machine learning. All of the research requires deep integration of data science, artificial intelligence, mathematics, and security. Students will learn to work in teams and communicate their results to a diverse audience by participating in activities that use investigative methodologies based on constructive dialogue and collaborative design. The projects have the potential to broaden knowledge in several domains, including game theory, intrusion detection systems, misinformation mitigation, and lightweight cryptography.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.