Gaussian Processes for Bayesian hypothesis tests on regression functions

Alessio Benavoli, Francesca Mangili
Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, PMLR 38:74-82, 2015.

Abstract

Gaussian processes have been used in different application domains such as classification, regression etc. In this paper we show that they can also be employed as a universal tool for developing a large variety of Bayesian statistical hypothesis tests for regression functions. In particular, we will use GPs for testing whether (i) two functions are equal; (ii) a function is monotone (even accounting for seasonality effects); (iii) a function is periodic; (iv) two functions are proportional. By simulation studies, we will show that, beside being more flexible, GP tests are also competitive in terms of performance with state-of-art algorithms.

Cite this Paper

BibTeX
@InProceedings{pmlr-v38-benavoli15, title = {{Gaussian Processes for Bayesian hypothesis tests on regression functions}}, author = {Benavoli, Alessio and Mangili, Francesca}, booktitle = {Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics}, pages = {74--82}, 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/benavoli15.pdf}, url = {https://proceedings.mlr.press/v38/benavoli15.html}, abstract = {Gaussian processes have been used in different application domains such as classification, regression etc. In this paper we show that they can also be employed as a universal tool for developing a large variety of Bayesian statistical hypothesis tests for regression functions. In particular, we will use GPs for testing whether (i) two functions are equal; (ii) a function is monotone (even accounting for seasonality effects); (iii) a function is periodic; (iv) two functions are proportional. By simulation studies, we will show that, beside being more flexible, GP tests are also competitive in terms of performance with state-of-art algorithms.} }
Endnote
%0 Conference Paper %T Gaussian Processes for Bayesian hypothesis tests on regression functions %A Alessio Benavoli %A Francesca Mangili %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-benavoli15 %I PMLR %P 74--82 %U https://proceedings.mlr.press/v38/benavoli15.html %V 38 %X Gaussian processes have been used in different application domains such as classification, regression etc. In this paper we show that they can also be employed as a universal tool for developing a large variety of Bayesian statistical hypothesis tests for regression functions. In particular, we will use GPs for testing whether (i) two functions are equal; (ii) a function is monotone (even accounting for seasonality effects); (iii) a function is periodic; (iv) two functions are proportional. By simulation studies, we will show that, beside being more flexible, GP tests are also competitive in terms of performance with state-of-art algorithms.
RIS
TY - CPAPER TI - Gaussian Processes for Bayesian hypothesis tests on regression functions AU - Alessio Benavoli AU - Francesca Mangili 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-benavoli15 PB - PMLR DP - Proceedings of Machine Learning Research VL - 38 SP - 74 EP - 82 L1 - http://proceedings.mlr.press/v38/benavoli15.pdf UR - https://proceedings.mlr.press/v38/benavoli15.html AB - Gaussian processes have been used in different application domains such as classification, regression etc. In this paper we show that they can also be employed as a universal tool for developing a large variety of Bayesian statistical hypothesis tests for regression functions. In particular, we will use GPs for testing whether (i) two functions are equal; (ii) a function is monotone (even accounting for seasonality effects); (iii) a function is periodic; (iv) two functions are proportional. By simulation studies, we will show that, beside being more flexible, GP tests are also competitive in terms of performance with state-of-art algorithms. ER -
APA
Benavoli, A. & Mangili, F.. (2015). Gaussian Processes for Bayesian hypothesis tests on regression functions. Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 38:74-82 Available from https://proceedings.mlr.press/v38/benavoli15.html.

Related Material