Changjian Shui
Ph.D.
Université Laval, Mila, Vector Institute
E-mail: changjian DOT shui AT vectorinstitute DOT ai
[Google Scholar] [Twitter] [CV]
About me
I am a postdoc fellow at the Vector Institute. Previously, I did my postdoc at McGill University/Mila, under the supervision of Prof. Tal Arbel. I obtained Ph.D. degree from Université Laval, under the supervision of Prof. Christian Gagné and Prof. Boyu Wang.
Research Interests I am working on a reliable and responsible use of Machine Learning. Specifically, I am intersted in machine learning under distribution shift by developing theories and algorithms. For example, algorithmic fairness, robustness, transfer learning, meta-learning, multitask learning, active learning.
Recent news [archive]
[Jan 2024] Our paper “Latent Trajectory Learning for Limited Timestamps under Distribution Shift over Time” is accepted at ICLR 2024 (Oral, 1.2% of the submission).
[Sep 2023] Our paper “On the Stability-Plasticity Dilemma in Continual Meta-Learning: Theory and Algorithm” is accepted at NeurIPS 2023.