Scalable Variational Gaussian Process Classification
Machine Learning
2014-11-10 v1
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.
Cite
@article{arxiv.1411.2005,
title = {Scalable Variational Gaussian Process Classification},
author = {James Hensman and Alex Matthews and Zoubin Ghahramani},
journal= {arXiv preprint arXiv:1411.2005},
year = {2014}
}
Comments
16 pages, 9 figures