معرفی
Trevor Campbell is an Associate Professor in the Department of Statistics at the University of British Columbia (UBC), Vancouver. He holds a Ph.D. in Machine Learning and Statistics from MIT and a B.A.Sc. in Aerospace Engineering from the University of Toronto. His research focuses on automated, scalable Bayesian inference algorithms, Bayesian nonparametrics, and streaming data analysis. Campbell is a core contributor to probabilistic programming tools like Pigeons.jl and has developed influential methods for coreset-based Bayesian inference.
Education:
- Ph.D. in Machine Learning and Statistics (2016), MIT
- M.S. in Aeronautics and Astronautics (2013), MIT
- B.A.Sc. in Aerospace Engineering (2011), University of Toronto
Research Interests:
His work emphasizes scalable Bayesian computation, including variational inference, MCMC optimization, and coresets. He develops algorithms that balance statistical accuracy with computational efficiency, particularly for large-scale datasets. His recent work explores adaptive samplers (e.g., AutoStep, autoMALA) and theoretical guarantees for coreset methods. Applications span astrophysics, materials science, and network analysis.
Awards & Grants:
- NSERC Discovery Grant (2025)
- Blackwell-Rosenbluth Award (2021)
- PIMS Early Career Award (2023)
- Google Perception Academic Funding (2020)
Advising & Labs:
He supervises a team of PhD and M.Sc. students at UBC, focusing on Bayesian methodology and computational tools. His lab collaborates with institutions like SFU and MIT on projects involving distributed sampling and probabilistic modeling. He co-organizes workshops on Bayesian computation and serves on editorial boards for Bayesian Analysis and TMLR.
