Mean field variational Bayesian inference for support vector machine classification
This work addresses the problem of improving SVM flexibility and performance for machine learning practitioners, though it is incremental as it builds on existing latent variable representations.
The paper tackles the limitations of classical support vector machines (SVMs) by introducing a mean field variational Bayesian inference method, which automatically selects penalty parameters, handles dependent samples, missing data, and variable selection, and outperforms classical SVMs on simulated and real datasets while remaining computationally efficient.
A mean field variational Bayes approach to support vector machines (SVMs) using the latent variable representation on Polson & Scott (2012) is presented. This representation allows circumvention of many of the shortcomings associated with classical SVMs including automatic penalty parameter selection, the ability to handle dependent samples, missing data and variable selection. We demonstrate on simulated and real datasets that our approach is easily extendable to non-standard situations and outperforms the classical SVM approach whilst remaining computationally efficient.