Bayesian Nonlinear Support Vector Machines for Big Data

Florian Wenzel, Théo Galy-Fajou, Matthäus Deutsch, Marius Kloft

ECML-PKDD 2017 pp. 307-322

doi:10.1007/978-3-319-71249-9_19 /ecmlpkdd/2017/wenzel2017ecmlpkdd-bayesian/

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.

PDF ECML-PKDD Semantic Scholar

Cite

Text

Wenzel et al. "Bayesian Nonlinear Support Vector Machines for Big Data." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2017. doi:10.1007/978-3-319-71249-9_19

Markdown

[Wenzel et al. "Bayesian Nonlinear Support Vector Machines for Big Data." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2017.](https://mlanthology.org/ecmlpkdd/2017/wenzel2017ecmlpkdd-bayesian/) doi:10.1007/978-3-319-71249-9_19

BibTeX

@inproceedings{wenzel2017ecmlpkdd-bayesian,
  title     = {{Bayesian Nonlinear Support Vector Machines for Big Data}},
  author    = {Wenzel, Florian and Galy-Fajou, Théo and Deutsch, Matthäus and Kloft, Marius},
  booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
  year      = {2017},
  pages     = {307-322},
  doi       = {10.1007/978-3-319-71249-9_19},
  url       = {https://mlanthology.org/ecmlpkdd/2017/wenzel2017ecmlpkdd-bayesian/}
}