PhD Student, Vanier Scholar
University of Toronto, Department of Statistical Sciences
Vector Institute
jeffrey.negrea AT mail.utoronto.ca
Biography
I’m a PhD candidate and Vanier Scholar at the University of Toronto Department of Statistical Sciences working on machine learning theory, online learning, and computational statistics.
Interests
- Online Learning
- Computational Statistics
- Statistical Learning Theory
- Applied Probability
- Markov Chains & Processes
Education
-
PhD in Statistical Sciences, (Ongoing)
University of Toronto
-
MSc in Statistical Sciences, 2017
University of Toronto
-
BMath Mathematical Finance, 2014
University of Waterloo
Publications
(2021).
Approximations of Geometrically Ergodic Reversible Markov Chains.
Advances in Applied Probability.
(2021).
Minimax Optimal Quantile and Semi-Adversarial Regret via Root-Logarithmic Regularizers.
Advances in Neural Information Processing Systems.
(2020).
In Defense of Uniform Convergence: Generalization via derandomization with an application to interpolating predictors.
Proceedings of the 37th International Conference on Machine Learning.
(2020).
Sharpened Generalization Bounds based on Conditional Mutual Information and an Application to Noisy, Iterative Algorithms.
Advances in Neural Information Processing Systems.
(2019).
Information-Theoretic Generalization Bounds for SGLD via Data-Dependent Estimates.
Advances in Neural Information Processing Systems.