Bayesian Multicategory Support Vector Machines
This work is incremental, offering a Bayesian reinterpretation of an existing method for multi-class classification tasks.
The paper tackles the problem of multi-class classification by providing a probabilistic interpretation of the multi-class support vector machine (MSVM) and extending it to a hierarchical Bayesian framework, showing that this approach maintains classification accuracy while offering Bayesian advantages.
We show that the multi-class support vector machine (MSVM) proposed by Lee et. al. (2004), can be viewed as a MAP estimation procedure under an appropriate probabilistic interpretation of the classifier. We also show that this interpretation can be extended to a hierarchical Bayesian architecture and to a fully-Bayesian inference procedure for multi-class classification based on data augmentation. We present empirical results that show that the advantages of the Bayesian formalism are obtained without a loss in classification accuracy.