Sparse Gaussian Processes Revisited: Bayesian Approaches to Inducing-Variable Approximations

Simone Rossi, Markus Heinonen, Edwin V. Bonilla, Zheyang Shen, Maurizio Filippone

Research output: Contribution to conferencePaperpeer-review

11 Scopus citations

Abstract

Variational inference techniques based on inducing variables provide an elegant framework for scalable posterior estimation in Gaussian process (gp) models. Besides enabling scalability, one of their main advantages over sparse approximations using direct marginal likelihood maximization is that they provide a robust alternative for point estimation of the inducing inputs, i.e. the location of the inducing variables. In this work we challenge the common wisdom that optimizing the inducing inputs in the variational framework yields optimal performance. We show that, by revisiting old model approximations such as the fully-independent training conditionals endowed with powerful sampling-based inference methods, treating both inducing locations and gp hyper-parameters in a Bayesian way can improve performance significantly. Based on stochastic gradient Hamiltonian Monte Carlo, we develop a fully Bayesian approach to scalable gp and deep gp models, and demonstrate its state-of-the-art performance through an extensive experimental campaign across several regression and classification problems.

Original languageEnglish (US)
Pages1837-1845
Number of pages9
StatePublished - 2021
Event24th International Conference on Artificial Intelligence and Statistics, AISTATS 2021 - Virtual, Online, United States
Duration: Apr 13 2021Apr 15 2021

Conference

Conference24th International Conference on Artificial Intelligence and Statistics, AISTATS 2021
Country/TerritoryUnited States
CityVirtual, Online
Period04/13/2104/15/21

ASJC Scopus subject areas

  • Artificial Intelligence
  • Software
  • Control and Systems Engineering
  • Statistics and Probability

Fingerprint

Dive into the research topics of 'Sparse Gaussian Processes Revisited: Bayesian Approaches to Inducing-Variable Approximations'. Together they form a unique fingerprint.

Cite this