Skip to yearly menu bar Skip to main content


Poster

Learning Gaussian Processes by Minimizing PAC-Bayesian Generalization Bounds

David Reeb · Andreas Doerr · Sebastian Gerwinn · Barbara Rakitsch

Room 210 #70

Keywords: [ Kernel Methods ] [ Learning Theory ] [ Gaussian Processes ] [ Regression ] [ Bayesian Theory ]


Abstract:

Gaussian Processes (GPs) are a generic modelling tool for supervised learning. While they have been successfully applied on large datasets, their use in safety-critical applications is hindered by the lack of good performance guarantees. To this end, we propose a method to learn GPs and their sparse approximations by directly optimizing a PAC-Bayesian bound on their generalization performance, instead of maximizing the marginal likelihood. Besides its theoretical appeal, we find in our evaluation that our learning method is robust and yields significantly better generalization guarantees than other common GP approaches on several regression benchmark datasets.

Live content is unavailable. Log in and register to view live content