Orthogonally Decoupled Variational Fourier Features
Machine Learning
2020-07-14 v1 Machine Learning
Abstract
Sparse inducing points have long been a standard method to fit Gaussian processes to big data. In the last few years, spectral methods that exploit approximations of the covariance kernel have shown to be competitive. In this work we exploit a recently introduced orthogonally decoupled variational basis to combine spectral methods and sparse inducing points methods. We show that the method is competitive with the state-of-the-art on synthetic and on real-world data.
Cite
@article{arxiv.2007.06363,
title = {Orthogonally Decoupled Variational Fourier Features},
author = {Dario Azzimonti and Manuel Schürch and Alessio Benavoli and Marco Zaffalon},
journal= {arXiv preprint arXiv:2007.06363},
year = {2020}
}