UNIVERSITY PARK, Pa. — The Penn State College of Information Sciences and Technology recently announced eight projects that will receive funding from the college’s seed grant program.
The program provides preliminary funding to support research activities and generate preliminary results that will eventually lead to bigger projects involving external funding. IST faculty wrote short proposals, which were reviewed by representatives of the college’s diverse faculty. The highest-rated proposals were recommended to IST’s dean and awarded on a competitive basis.
“These projects demonstrate the interdisciplinary research that is a hallmark of the College of IST,” said Andrea Tapia, associate dean for research in IST. “They explore important societal challenges that can be addressed through information and technology and show tremendous potential for future success.”
The eight projects selected for funding are:
- “AI Learning to Retrieve and Generate Educational Videos with Equitable Ambient Cues” by Sharon X. Huang, professor of IST, and ChanMin Kim, associate professor in the College of Education. This project aims to explore the use of ambient cues — objects that are not discussed in foreground conversation but signal who belongs and who does not — in elementary school STEM classrooms that are aligned with identities of racially and/or ethnically minority students for improving equity and inclusion. The researchers will develop artificial intelligence techniques to help teachers find video materials with ambient cues in association with learner identities such as race, ethnicity, gender and age.
- “How far are we in developing sustainable ecosystem: AI-based Wildlife data analytics” by Johnson Kinyua, associate teaching professor of IST; Shreya Ghosh, postdoctoral scholar in IST; Prasenjit Mitra, professor of IST; and Titus Adhola, Wildlife Management and Conservation, Department of Clinical Studies, University of Nairobi, Kenya. The project will investigate signature movement and human wildlife conflict patterns of different taxonomy of animals and how their movement habits change over spatial and temporal features, including season, day/night, and spatial variables. The researchers aim to develop a systematic wildlife analytics framework for data sharing, extracting insights and utilizing the knowledge to build a sustainable ecosystem.
- “Human-Centered Risk Assessment of Period and Fertility Tracking Post Roe v. Wade” by Xinning Gui, assistant professor of IST, and Yao Li, assistant professor at the University of Central Florida. Many people in the U.S. use period and fertility tracking apps for reproductive health care, and they are concerned about the risks of using these apps. This project seeks to use a human-centered design approach to investigate privacy practices and perceptions from the perspectives of both end users and developers of period and fertility tracking apps to understand the implications for design and policy-making in the post-Roe era.
- “Mapping Transnational Histories of Surveillance Technologies, Expertise, and Institutional Exchanges between Southeast Asia and the Appalachian United States” by Cindy Lin, assistant professor of IST, and Andrea Miller, assistant professor of telecommunications and women's, gender, and sexuality studies in the College of the Liberal Arts. This project will map and examine how longstanding transnational exchanges and circulations of surveillance technology and technoscientific expertise between Southeast Asia and the Appalachian United States have influenced contemporary environmental concerns and governance projects.
- “Robot Firmware Semantic Recovery and Vulnerability Discovery” by Taegyu Kim, assistant professor of IST, and Dinghao Wu, professor of IST. Despite robots increasing deployment for various tasks, these systems also have numerous issues related to cyber and physical components and their interplay. This project aims to mitigate attacks on these systems through control module semantic recovery and cyber-physical vulnerability discovery techniques.
- “Robust Hyperparameter Optimization Under Concept Drift” by Qingyun Wu, assistant professor of IST. The project aims to identify machine learning models that can effectively handle concept drifts via hyperparameter optimization (HPO). The research strives to gain a better understanding of how hyperparameters impact the out-of-distribution generalization of machine learning models and to develop more effective HPO methods for building robust models that can generalize well to new data.
- “Teaching Basic Chemistry to Students at Risk” by Frank Ritter, professor of IST, Serhii Serdiuk, associate research professor of IST, and Mary Jo Bojan, teaching professor in the Eberly College of Science. The researchers aim to create a tutorial system to teach foundational chemistry skills to students at risk in their first university chemistry course. The resulting tutorials will be used to help students in introductory chemistry courses at Penn State, with the results used to propose broader educational interventions.
- “Towards Adversarially Robust Anomaly Detection Through Diffusion Model” by Lu Lin, assistant professor of IST. This research aims to provide a rigorous adversarial attack formulation for anomaly detection, evaluate the risk of deep anomaly detectors, and build a principled adversarially robust anomaly detection method to mitigate such risks.