Research & Publications

About My Research

I do research in mathematics, in particular foundations of data science and machine learning. Lately, I am interested in three directions:

  • Optimization for Machine Learning
  • Online Learning, Learning in Games & Reinforcement Learning Theory
  • Modern Paradigms in Learning & Deep Learning Theory

On the side, I also work on interactive learning, privacy (differential privacy), and robustness. Beyond these areas, I also enjoy working on problems of independent interest, currently in number theory, combinatorics, and discrete geometry.

You can also find my publications on my Google Scholar profile.

If you are interested in collaborating, I’d love to hear from you -- reach out at viverson@uwaterloo.ca!

Publications & Manuscripts

Valentio Iverson, Gautam Kamath, Argyris Mouzakis
Accepted to Conference on Learning Theory (COLT) 2025 and Theory and Practice of Differential Privacy (TPDP) 2025
A Pathological Property of Nonlocal Discrete Operators
William Chang, Christopher Goodrich, Valentio Iverson, Khang Nguyen
Accepted to Proceedings of the American Mathematical Society (PAMS)
Accepted to International Conference on Acoustics, Speech, and Signal Processing (ICASSP) 2025

Talks

Valentio Iverson
University of Waterloo Student Number Theory Seminar, 2023

I am deeply grateful to my mentors, collaborators, and friends who have profoundly shaped my path as a researcher. I extend particular gratitude to Xiaoheng Wang for introducing me to math research, Stephen Vavasis and Gautam Kamath for being the first ones to mentor me and involve me in interesting machine learning theory research, and Argyris Mouzakis for his invaluable mentorship and discussions throughout the year. I am also indebted to Shai Ben-David and Aukosh Jagannath for my interest in this field.