Sort, Partition, Randomize: Optimal Binary Hypothesis Testing under Local Differential Privacy arXiv
Selected for Lightning Talk at Topology, Algebra, and Geometry in Data Science (TAG-DS) 2026
Elena Ghazi
I am a PhD student in Computer Science at Harvard, where I am very fortunate to be advised by Flavio Calmon and Salil Vadhan. My research interests include privacy-preserving AI, AI for health, and their intersection.
In summer 2026, I am interning at Boston Children's Hospital's Computational Health Informatics Program (CHIP), working on AI for health with Scott Wang and Michael Lingzhi Li.
I am part of Harvard's Information Theory Lab and Theory of Computation Group, and the OpenDP open-source software project.
Before joining Harvard, I was a software engineer at Google, working on privacy and compliance for wearables and health data.
Check out the Differential Privacy for Health and Genomics workshop that will be held on June 2–3, 2026 in Boston, MA!
Recent updates
- Co-organizing the Differential Privacy for Health and Genomics workshop.
- Co-presented a talk and two posters at Theory and Practice of Differential Privacy.
- Differential Privacy Deployments Registry featured by Harvard SEAS News.
- Paper accepted to the IEEE Symposium on Security and Privacy.
- Differential Privacy Deployments Registry launch announced by OpenDP.
- Best Poster/Demo at NYC Privacy Day for the deployment registry project.
Selected papers and projects
Making Privacy Public: Toward a Differential Privacy Deployment Registry IEEE S&P arXiv
2026 IEEE Symposium on Security and Privacy
Best Poster/Demo at NYC Privacy Day 2025
Featured
Harvard SEAS News Privacy Goes Public With New Database News story on the Differential Privacy Deployments Registry and the Harvard research behind it. March 30, 2026 OpenDP Blog Launching the Differential Privacy Deployments Registry Project announcement covering the registry launch, NIST proposal, and user-study research. November 25, 2025A Community-Driven Differential Privacy Deployment Registry NIST IR 8588 (Initial Public Draft)
Contributed Talk at Theory and Practice of Differential Privacy 2026
On User-Level Differential Privacy and Timing Attacks
Total Variation Meets Differential Privacy IEEE JSAIT arXiv
2024 IEEE Journal on Selected Areas in Information Theory (JSAIT)
Total Variation with Differential Privacy: Tighter Composition and Asymptotic Bounds IEEE ISIT 2023
2023 IEEE International Symposium on Information Theory (ISIT)
