Me

Argyris Mouzakis

PhD student
Cheriton School of Computer Science
University of Waterloo
Office: DC 2306N1
200 University Ave W Waterloo, ON N2L 3G1
amouzaki@uwaterloo.ca

Hello! I'm a fifth year PhD student at the University of Waterloo's Cheriton School of Computer Science, where I'm very fortunate to be a member of The Salon, and be advised by Gautam Kamath. My interests broadly lie at the intersection of Theoretical Computer Science and Statistics, with a particular focus on Machine Learning Theory, Algorithmic Statistics, and Differential Privacy. Before joining UWaterloo, I did my undergrad in ECE at NTUA, where I was a member of CoReLab, and was advised by Dimitris Fotakis. [CV] [Last Updated: Oct. 2024]
Here's a tutorial about how to pronounce my name. If it's hard for you, I'm happy with people calling me Argy (RG). :)

Pronouns: I identify as a cisgender man, but I am fine with anything respectful.

Since September 2023, I am a proud recipient of a scholarship from the Onassis Foundation, for the support of my doctoral studies!

I help coordinate the A&C group's seminar talks at Waterloo. If you're interested in giving a talk, feel free to e-mail me!


Publications and Manuscripts

For the following list, author names are in alphabetical order, as is customary in TCS (unless clearly stated otherwise). Depending on when you're looking, the list may or may not be up-to-date, so here's my (more likely to be up-to-date) DBLP and Google Scholar profiles.

Private Mean Estimation with Person-Level Differential Privacy

SODA 2025

Not All Learnable Distribution Classes are Privately Learnable

ALT 2024

A Bias-Variance-Privacy Trilemma for Statistical Estimation

TPDP 2023

New Lower Bounds for Private Estimation and a Generalized Fingerprinting Lemma

Gautam Kamath, Argyris Mouzakis, Vikrant Singhal
NeurIPS 2022; TPDP 2022

A Private and Computationally Efficient Estimator for Unbounded Gaussians

COLT 2022; TPDP 2022