Thursday, February 26, 2009

The Infinite Gaussian Mixture Model


This is an even earlier paper on DP. Rasmussen employs CRP though, a little different from the SBP. He first studies the finite mixture model. Then take the number of components to infinity and modify the mixing proportion correspondingly. The inference then is done by MCMC. I am not quite clear why he sets up such a complicated model -,-

No comments: