Variational Gaussian Processes: A Functional Analysis View
This work addresses a methodological gap for researchers in machine learning, offering a more general framework for variational Gaussian process approximations, though it appears incremental as it builds on existing techniques.
The paper tackles the lack of generality in selecting variational features for fast Gaussian process inference by proposing a Banach space view, which unifies existing features and connects kernel ridge regression to variational approximations.
Variational Gaussian process (GP) approximations have become a standard tool in fast GP inference. This technique requires a user to select variational features to increase efficiency. So far the common choices in the literature are disparate and lacking generality. We propose to view the GP as lying in a Banach space which then facilitates a unified perspective. This is used to understand the relationship between existing features and to draw a connection between kernel ridge regression and variational GP approximations.