Dr. Daniel Roy

Biography

Daniel M. Roy is an Associate Professor of Statistics at the University of Toronto, with cross-appointments in Computer Science and in Electrical and Computer Engineering. His research focuses on foundational principles in prediction, inference, and decision-making under uncertainty, spanning machine learning, statistics, mathematical logic, applied probability, and computer science. His work includes significant contributions to learning theory, statistical network analysis, decision theory, probabilistic programming, and Bayesian nonparametric statistics. Roy’s research in deep learning spans theory and practice. His contributions range from his pioneering work on empirically grounded statistical theory for deep learning, to state-of-the-art algorithms for neural network compression and data-parallel training. His experimental work has shed light on various deep learning phenomena, including neural network training dynamics and linear mode connectivity, while his recent theoretical work introduces simple but accurate mathematical models for deep neural networks at initialization. Beyond his contributions to deep learning, Roy has made significant advances to the mathematical and statistical underpinnings of AI. His dissertation on probabilistic programming languages and computable probability theory was recognized by an MIT Sprowls Award. Roy recently resolved several open problems in statistical decision theory posed over 70 years ago, by exploiting the properties of infinitesimal numbers to expand the set of allowable Bayesian priors. His latest work, focussing on robust and adaptive decision-making, has been recognized by multiple oral presentations at leading conferences and best poster awards. Undergraduates at Toronto: I welcome these emails, but they should include a CV, lists of relevant coursework, and a transcript. The subject line should contain the word "consideration" to indicate these instructions have been read. Students seeking short-term research opportunities: Other than in exceptional cases (listed in the n paragraph), I do not take students for internships, summer research, or short-term visits. Prospective graduate students: I receive emails every day from prospective graduate students. Other than in exceptional cases, students who are not already admitted to Toronto should NOT email me directly. Instead, they should first apply to and gain admission to the graduate program in Statistics or Computer Science. Do not email me about

Research Interests

Courses