MDL cost function for GMM model selection

I used this function when I was working on the paper about online incremental learning two years ago. But cannot recall it when someone asked me on this recently.  Put it here to memorize it:
where big theta is the model (represented by a set of parameters). The total number of parameter of the GMM is N(k), and k is the number of Gaussian components. y(obs) are the observations.  n is the number of observations.
Advertisements
This entry was posted in Research. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s