Bayesian Nonlinear Support Vector Machines for Big Data

Florian Wenzel,Theo Galy-Fajou,Matthaeus Deutsch,Marius Kloft
DOI: https://doi.org/10.48550/arXiv.1707.05532
IF: 5.414
2017-07-18
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.
What problem does this paper attempt to address?