Scalable variational Gaussian process classification

Hensman, James and Matthews, Alexander G. and Ghahramani, Zoubin (2015) Scalable variational Gaussian process classification. Proceedings of Machine Learning Research, 38. pp. 351-360. ISSN 1938-7228

Abstract

Gaussian process classification is a popular method with a number of appealing properties. We show how to scale the model within a variational inducing point framework, outperforming the state of the art on benchmark datasets. Importantly, the variational formulation can be exploited to allow classification in problems with millions of data points, as we demonstrate in experiments.