Dayi (David) Li (李大一), Ph.D.
NSERC Canada Postdoctoral Research Award Fellow, University of Waterloo
Enabling Efficient and Robust Inference for Complex Scientific Problems
I am an NSERC Canada Postdoctoral Research Award Fellow at the University of Waterloo, where I work with Alex Stringer and Will Percival. Before joining Waterloo, I was a postdoctoral fellow and PhD student at the University of Toronto, where I worked with Gwendolyn Eadie, Patrick Brown, and Roberto Abraham. During my doctoral studies, I was a Data Sciences Institute Doctoral Fellow and a CANSSI Ontario Multidisciplinary Doctoral trainee.
My research focuses on developing efficient and robust surrogate-based Bayesian computational methods for complex, computationally expensive scientific models arising in cosmology and astrophysics. On the theoretical side, I study the statistical convergence guarantees of these methods, together with practical measures of their performance and reliability.
I am also interested in developing statistical models for extracting scientifically meaningful signals from complex data. One example is the detection of ultra-diffuse galaxies through the spatial distribution of their globular clusters.
July 2026 — My first single-authored paper, "Robust Surrogate-Based Bayesian Inference via Sampling-Based Adaptive Active Learning (SALE)" is now out on arXiv.
Mar 2026 — I will be joining the University of Waterloo as a Canada Postdoctoral Research Award Fellow, jointly appointed between the Department of Statistics and Actuarial Science and the Waterloo Centre for Astrophysics.
Feb 2026 — Check out the NASA news coverage on our discovery of Candidate Dark Galaxy-2.