David Huk
I am a third-year PhD student in the CDT in Mathematics and Statistics at Warwick. I am jointly supervised by Dr Rito Dutta and Professor Mark Steel. My main research interests are computational statistics, multivariate dependence, and reinforcement learning.
Recently, my work on Quasi-bayesian vines has been accepted at Neurips 24, and I have freshly finished a second paper identifying a link between copulas and classification. Currently, I focus on applying these ideas within reinforcement learning to enhance agents.
Publications:
- Your copula is a classifier in disguise: classification-based copula density estimation, PreprintLink opens in a new window
David Huk, Mark Steel, and Ritabrata Dutta
- Quasi-Bayes meets Vines, Neurips (2024)Link opens in a new window
David Huk, Yuanhe Zhang, Mark Steel, Ritabrata Dutta
Link opens in a new window
Recorded 5 minute video: https://youtu.be/6kTmdM7DUUELink opens in a new window
- Contribution to the Discussion of “Martingale Posterior Distributions”, Journal of the Royal Statistical Society Series B: Statistical Methodology (2023)Link opens in a new window
David Huk, Lorenzo Pacchiardi, Ritabrata Dutta and Mark Steel
Link opens in a new window - Probabilistic Rainfall Downscaling: Joint Generalized Neural Models with Censored Spatial Gaussian Copula, Preprint (2023)Link opens in a new window
David Huk, Rilwan A. Adewoyin, Ritabrata Dutta
Link opens in a new window
Talks and Presentations
- Quasi-Bayes meets Vines, Colloquium of the Department of Mathematics, Technical University of Munich, invited talk, (January 2025)
- Quasi-Bayes meets Vines, The Thirty-Eighth Annual Conference on Neural Information Processing Systems, Neurips 24, Vancouver, poster, (December 2024)
- Quasi-Bayes meets Vines, Algorithms & Computationally Intensive Inference seminar, University of Warwick, invited talk, (June 2024)
- Probabilistic Rainfall Downscaling: Joint Generalized Neural Models with Censored Spatial Gaussian Copula,
European Centre for Medium-Range Weather Forecasts, online, invited talk (October 2023) - Probabilistic forecasting with censored spatial copulas via scoring rules, CRiSM Workshop on Fusing Simulations with Data Science, University of Warwick, poster (July 2023)
- Probabilistic forecasting with censored spatial copulas via scoring rules, Workshop on Distance-based Methods in Machine Learning, University College London, poster (June 2023)
- Joint Generalized Neural Models and Censored Spatial Copulas for Probabilistic Rainfall Forecasting, European Geosciences Union General Assembly 2023, Vienna, poster (April 2023)
Academic Roles
Chair of Student-Staff Liaison Committee in Statistics, 24/25
Student-Staff Liaison Committee member in Statistics, 23/24 and 24/25
Research Committee member in Statistics, 24/25
Tutorial Teaching
- ST 404 Applied Statistical Modelling (Term 2, 2023/2024)
- ST 219 Mathematical Statistics Part B (Term 2, 2022/2023)
