Hierarchical Inducing Point Gaussian Process for Inter-domian Observations

Luhuan Wu · Andrew Miller · Lauren Anderson · Geoff Pleiss · David Blei · John Cunningham

Keywords: [ Models and Methods ] [ Gaussian Processes ]

[ Abstract ]
Tue 13 Apr 6:30 p.m. PDT — 8:30 p.m. PDT


We examine the general problem of inter-domain Gaussian Processes (GPs): problems where the GP realization and the noisy observations of that realization lie on different domains. When the mapping between those domains is linear, such as integration or diļ¬€erentiation, inference is still closed form. However, many of the scaling and approximation techniques that our community has developed do not apply to this setting. In this work, we introduce the hierarchical inducing point GP (HIP-GP), a scalable inter-domain GP inference method that enables us to improve the approximation accuracy by increasing the number of inducing points to the millions. HIP-GP, which relies on inducing points with grid structure and a stationary kernel assumption, is suitable for low-dimensional problems. In developing HIP-GP, we introduce (1) a fast whitening strategy, and (2) a novel preconditioner for conjugate gradients which can be helpful in general GP settings.

Chat is not available.