Project Grant 2350333

The University of Virginia (UVA) received a $197,290 Project Grant award from the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) to conduct collaborative research on security and privacy in machine unlearning. The project aims to better understand and defend against security and privacy risks that may arise from the use of machine unlearning techniques, which allow companies to remove the influence of personal data from their machine learning models. The research is organized around three main thrusts: (1) investigating backdoor and model stealing attacks that exploit the unlearning process, (2) designing enhanced privacy-centric attacks like membership inference and data reconstruction, and (3) developing strategies to detect malicious unlearning requests and improve model resilience. The broader impacts of this work include transferring technologies to industry, increasing underrepresented group involvement in computing research, and disseminating outcomes through K-12 outreach and community services. This award is jointly funded by the Secure and Trustworthy Cyberspace (SATC) program and the Established Program to Stimulate Competitive Research (EPSCoR).