Print Email Facebook Twitter Know Your Boundaries Title Know Your Boundaries: Constraining Gaussian Processes by Variational Harmonic Features Author Solin, Arno (Aalto University) Kok, M. (TU Delft Team Jan-Willem van Wingerden) Contributor Chaudhuri, Kamalika (editor) Sugiyama, Masashi (editor) Date 2019 Abstract Gaussian processes (GPs) provide a powerful framework for extrapolation, interpolation, and noise removal in regression and classification. This paper considers constraining GPs to arbitrarily-shaped domains with boundary conditions. We solve a Fourier-like generalised harmonic feature representation of the GP prior in the domain of interest, which both constrains the GP and attains a lowrank representation that is used for speeding up inference. The method scales as O(nm2) in prediction and O(m3) in hyperparameter learning for regression, where n is the number of data points and m the number of features.Furthermore, we make use of the variational approach to allow the method to deal with non-Gaussian likelihoods. The experiments cover both simulated and empirical data in which the boundary conditions allow for inclusion of additional physical information. To reference this document use: http://resolver.tudelft.nl/uuid:86de4906-4d76-412c-8ce3-e1e8ad7ed56f Publisher MLR Press Source Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics (AISTATS 2019) Event AISTATS 2019 2019: 22nd International Conference on Artificial Intelligence and Statistics, 2019-04-16 → 2019-04-18, Naha, Okinawa, Japan Series Proceedings of Machine Learning Research (PMLR), 2640-3498, 89 Part of collection Institutional Repository Document type conference paper Rights © 2019 Arno Solin, M. Kok Files PDF article.pdf 1.72 MB Close viewer /islandora/object/uuid:86de4906-4d76-412c-8ce3-e1e8ad7ed56f/datastream/OBJ/view