AISTATS 2022poster6 citations

Variational Gaussian Processes: A Functional Analysis View

George Wynne, Veit Wild

Abstract

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.

BibTeX
@InProceedings{pmlr-v151-wynne22a,
  title = 	 { Variational Gaussian Processes: A Functional Analysis View },
  author =       {Wynne, George and Wild, Veit},
  booktitle = 	 {Proceedings of The 25th International Conference on Artificial Intelligence and Statistics},
  pages = 	 {4955--4971},
  year = 	 {2022},
  editor = 	 {Camps-Valls, Gustau and Ruiz, Francisco J. R. and Valera, Isabel},
  volume = 	 {151},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {28--30 Mar},
  publisher =    {PMLR},
  pdf = 	 {https://proceedings.mlr.press/v151/wynne22a/wynne22a.pdf},
  url = 	 {https://proceedings.mlr.press/v151/wynne22a.html},
  abstract = 	 { 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. }
}
Variational Gaussian Processes: A Functional Analysis View · AISTATS 2022