Poster
in
Workshop: Gaussian Processes, Spatiotemporal Modeling, and Decision-making Systems
Integrated Fourier Features for Fast Sparse Variational Gaussian Process Regression
Talay Cheema
Abstract:
Sparse variational approximations are popular methods for scaling up inference in Gaussian processes to larger datasets. For training points, exact inference has cost; with features, sparse variational methods have cost. Recently, methods have been proposed using harmonic features; when the domain is spherical, the resultant method has cost, but in the common case of a Euclidean domain, previous methods do not avoid the scaling and are generally limited to a fairly small class of kernels. In this work, we propose integrated Fourier features, with which we can obtain cost, and the method can easily be applied to any covariance function for which we can easily evaluate the spectral density. We provide convergence results, and synthetic experiments showing practical performance gains.
Chat is not available.