The ability to collect and analyze datasets has tremendous benefits to society. However, such benefits must not come at the price of privacy. This calls for secure and efficient methods for computing on private data that can resolve the conflict between the benefits of computation and the risk of privacy loss.The powerful paradigm of secure multiparty computation (MPC) can enable computation over datasets of mutually distrusting individuals and organizations while still preserving their privacy. Significant research advances in recent years have brought MPC closer than ever to practice. Nevertheless, many existing and emerging applications demand new efficiency and resiliency properties that are outside the reach of known solutions.
To meet these demands, this project initiates two new lines of research in the study of MPC. First, to answer calls for MPC-as-a-service and MPC deployments in volunteer-based networks, this project develops new methods to enable large-scale computations. Second, to improve the resilience of MPC in strongly adversarial environments, this project develops new models and techniques to prevent data exfiltration even in the face of tampering attacks over protocol communication and computation. The project has outreach activities to raise awareness about security and cryptography in the greater Baltimore area.
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.
