Information Security Group --- Computer Science Department --- University College London
Research interests: anonymous communication (especially low-latency AC protocols, such as Tor), privacy metrics (especially metrics based on differential privacy), machine learning (especially in a privacy-preserving fashion)
I'm currently studying how to quantify (differential) privacy in a tight and lossless fashion to maximize the trade-off between privacy and utility. I find this quite interesting, since tighter bounds, especially under composition, can be applied to improve a variety of existing privacy-preserving mechanisms, e.g., modern machine learning methods that often operate on sensitive data in a heavily repeated fashion.
All Anonymity Differential Privacy Cryptography