Project Image Mixture Model MCMC Inference
We develop parallelizable samplers for Dirichlet process mixture models that do not require approximating the infinite model. Two sub-clusters are fit for each regular-cluster, and are used to propose large split and merge moves. Inference is shown to be orders of magnitude faster than traditional Gibbs sampling while being more robust to different initializations and hyper-parameters.

People Involved: Jason Chang, John W. Fisher III

Code for our DPMM sampler can be found here.

In the following video, we show an illustrative example of the sub-cluster sampling splits. The two inferred sub-cluster parameters are shown with ellipses. Color correspond to the cluster assignment.