English

Bayesian Nonlinear Support Vector Machines for Big Data

Machine Learning 2018-03-22 v1 Machine Learning

Abstract

We propose a fast inference method for Bayesian nonlinear support vector machines that leverages stochastic variational inference and inducing points. Our experiments show that the proposed method is faster than competing Bayesian approaches and scales easily to millions of data points. It provides additional features over frequentist competitors such as accurate predictive uncertainty estimates and automatic hyperparameter search.

Keywords

Cite

@article{arxiv.1707.05532,
  title  = {Bayesian Nonlinear Support Vector Machines for Big Data},
  author = {Florian Wenzel and Theo Galy-Fajou and Matthaeus Deutsch and Marius Kloft},
  journal= {arXiv preprint arXiv:1707.05532},
  year   = {2018}
}

Comments

accepted as conference paper at ECML-PKDD 2017