I will present our work on interpeting the error of private median queries at TPDP 2026 in Boston.
June 2025I started as a research intern at MSR Redmond with Zinan Lin!
May 2025Our paper FastLloyd: Federated, Accurate, Secure, and Tunable k-Means Clustering with Differential Privacy is accepted to USENIX Security 2025!
I am currently a PhD Candidate at the University of Waterloo under the supervision of Florian Kerschbaum, working in the Cryptography Security and Privacy (CrySP) Lab.
I received my Master’s Degree in Computer Science from the University of Waterloo with Florian Kerschbaum. Before that I obtained my Bachelor of Science Degree with a double major in Mathematics and Computer Science from Brandon University. I have interned with Microsoft Research twice in both Cambridge and Redmond.
My research focuses on preserving privacy in machine learning, artificial intelligence, and data science. Specifically, enabling the deployment of systems with meaningful privacy trade-offs through strategic algorithmic design and careful auditing to expose privacy oversights. My research spans the fields of evolutionary algorithms, differential privacy, and secure computation. To learn more about my research, check out my publications page!
© 2026 Thomas Humphries
Website template by Spencer H. Bryngelson