Dino Sejdinovic: Recent Developments at the Interface Between Kernel Embeddings and Gaussian Processes
Time: Tue 2021-10-19 10.15
Lecturer: Dino Sejdinovic (University of Oxford)
Reproducing kernel Hilbert spaces (RKHS) provide a powerful framework, termed kernel mean embeddings, for representing probability distributions, enabling nonparametric statistical inference in a variety of applications. I will give an overview of this framework and present some of its recent developments which combine RKHS formalism with Gaussian process modeling. Some recent applications include causal data fusion, where data of different quality needs to be combined in order to estimate the average treatment effect, as well as statistical downscaling using potentially unmatched multi-resolution data.
S. L. Chau, S. Bouabid, and D. Sejdinovic, Deconditional Downscaling with Gaussian Processes, in Advances in Neural Information Processing Systems (NeurIPS), 2021, forthcoming. arXiv:2105.12909
S. L. Chau, J.-F. Ton, J. Gonzalez, Y. W. Teh, and D. Sejdinovic, BayesIMP: Uncertainty Quantification for Causal Data Fusion, in Advances in Neural Information Processing Systems (NeurIPS), 2021, forthcoming. arXiv:2106.03477