Scalable Variational Gaussian Process Classification

James Hensman, Alexander Matthews, Zoubin Ghahramani
Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, PMLR 38:351-360, 2015.

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, out-performing the state of the art on benchmark datasets. Importantly, the variational formulation an be exploited to allow classification in problems with millions of data points, as we demonstrate in experiments.

Cite this Paper


BibTeX
@InProceedings{pmlr-v38-hensman15, title = {{Scalable Variational Gaussian Process Classification}}, author = {Hensman, James and Matthews, Alexander and Ghahramani, Zoubin}, booktitle = {Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics}, pages = {351--360}, year = {2015}, editor = {Lebanon, Guy and Vishwanathan, S. V. N.}, volume = {38}, series = {Proceedings of Machine Learning Research}, address = {San Diego, California, USA}, month = {09--12 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v38/hensman15.pdf}, url = {https://proceedings.mlr.press/v38/hensman15.html}, 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, out-performing the state of the art on benchmark datasets. Importantly, the variational formulation an be exploited to allow classification in problems with millions of data points, as we demonstrate in experiments.} }
Endnote
%0 Conference Paper %T Scalable Variational Gaussian Process Classification %A James Hensman %A Alexander Matthews %A Zoubin Ghahramani %B Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2015 %E Guy Lebanon %E S. V. N. Vishwanathan %F pmlr-v38-hensman15 %I PMLR %P 351--360 %U https://proceedings.mlr.press/v38/hensman15.html %V 38 %X 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, out-performing the state of the art on benchmark datasets. Importantly, the variational formulation an be exploited to allow classification in problems with millions of data points, as we demonstrate in experiments.
RIS
TY - CPAPER TI - Scalable Variational Gaussian Process Classification AU - James Hensman AU - Alexander Matthews AU - Zoubin Ghahramani BT - Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics DA - 2015/02/21 ED - Guy Lebanon ED - S. V. N. Vishwanathan ID - pmlr-v38-hensman15 PB - PMLR DP - Proceedings of Machine Learning Research VL - 38 SP - 351 EP - 360 L1 - http://proceedings.mlr.press/v38/hensman15.pdf UR - https://proceedings.mlr.press/v38/hensman15.html AB - 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, out-performing the state of the art on benchmark datasets. Importantly, the variational formulation an be exploited to allow classification in problems with millions of data points, as we demonstrate in experiments. ER -
APA
Hensman, J., Matthews, A. & Ghahramani, Z.. (2015). Scalable Variational Gaussian Process Classification. Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 38:351-360 Available from https://proceedings.mlr.press/v38/hensman15.html.

Related Material