Homepage of Viet Nguyen

I'm a PhD student under the supervision of prof. Rahul G. Krishnan at the Vector Institute and in the Dept. of Computer Science at the University of Toronto. Broadly, I am interested in intelligence, biological and artificial. My primary research areas are:

• Compositional generalization, applications in vision-language models
• Model deterioration monitoring, guardrails for AI
• Meta-learning, in-context learning, chain-of-thought

Previously, I earned a B.Sc. in Probability and Statistics and a M.Sc. in Computer Science at Mila and McGill under the supervision of prof. Doina Precup where I worked on theoretical reinforcement learning, supported by the DeepMind Graduate Award.

Email: viet at cs dot toronto dot edu

Latest News

• [May. 1, 2025] Our paper "Diverse Prototypical Ensembles Improve Robustness to Subpopulation Shift" got accepted to ICML 2025!
• [Mar. 24, 2025] I presented our work "Post-Deployment Deterioration Monitoring" at the 2nd Vector Institute & RIKEN AIP Joint Symposium on Machine Learning and Artificial Intelligence!
• [Feb. 14, 2025] Our package bayesian_dpddm implementing Bayesian D-PDDM for post-deployment deterioration monitoring of ML models is released on PyPI!
• [Dec. 31, 2024] I got a film camera!
• [Sep. 26, 2024] Our paper "Sequential Decision Making with Expert Demonstrations under Unobserved Heterogeneity" got accepted to NeurIPS 2024!

Useful links

• GitHub
• Geometric Analysis Seminar
• nLab
• The Centre de recherches mathématiques (CRM)
• Vector Support
• torch wheels
• Matrix Cookbook
• The Stacks project
• Nguyen Ngoc Tu's writings

Cooler but less useful links

• Google Scholar
• Friends
• Projects
• CV
• Master's thesis
• Logs