-
Physics-Informed Variational State-Space Gaussian ProcessesOliver Hamelijnck*, Arno Solin, Theodoros DamoulasTo appear at the Thirty-Eighth Conference on Neural Information Processing Systems (NeurIPS 2024)[paper]
Biography
I am a postdoctoral researcher at the University of Warwick, working on the Machine Learning Foundations of Digital Twins project under the supervision of Theodoros Damoulas. My research focuses on Machine Learning and spatio-temporal Statistics, particularly scalable inference methods for non-parametric Bayesian models such as Gaussian Processes. I am also interested in multi-fidelity and multi-task modeling, with a focus on integrating structure through non-stationary, non-separable priors and physical laws.