Carnegie Mellon University
file.pdf (3.38 MB)

Infinite SVM: a Dirichlet Process Mixture of Large-margin Kernel Machines

Download (3.38 MB)
journal contribution
posted on 2011-06-01, 00:00 authored by Jun Zhu, Ning Chen, Eric P Xing

We present Infinite SVM (iSVM), a Dirichlet process mixture of large-margin kernel machines for multi-way classification. An iSVM enjoys the advantages of both Bayesian nonparametrics in handling the unknown number of mixing components, and large-margin kernel machines in robustly capturing local nonlinearity of complex data. We develop an efficient variational learning algorithm for posterior inference of iSVM, and we demonstrate the advantages of iSVM over Dirichlet process mixture of generalized linear models and other benchmarks on both synthetic and real Flickr image classification datasets.


Publisher Statement

Copyright 2011 by the author(s)/owner(s)



Usage metrics