9
$\begingroup$

In the context of online clustering, I often find many papers talking about: "dirichlet process" and "finite/infinite mixture models".

Given that I've never used or read about dirichlet process or mixture models. Do you know any suggestions of introductory lectures or papers that are easy to understand, about that ?

$\endgroup$
3
  • $\begingroup$ Is your background in statistics or in computer science? $\endgroup$
    – MånsT
    Commented Jun 15, 2012 at 11:44
  • $\begingroup$ @MånsT I know about classical probability theory and the usual distributions (gaussian etc), I've already designed and implemented some online clustering algorithms, and wanted recently to make use of particle filters to do some online clustering, but each time I find papers talking about DP and Mixture models, so I want to know about that. $\endgroup$
    – shn
    Commented Jun 15, 2012 at 11:54
  • $\begingroup$ I like Finite mixture models by McLachlan and Peel. I haven't really got any recommendations for DP though. $\endgroup$
    – MånsT
    Commented Jun 15, 2012 at 11:58

3 Answers 3

8
$\begingroup$

This is a gentle tutorial:

http://www.cs.cmu.edu/~kbe/dp_tutorial.pdf

I like the wikipedia entry as well. The links there are also very good:

http://en.wikipedia.org/wiki/Dirichlet_process

Here is a summer school lecture of one of the most active researchers

http://videolectures.net/mlss07_teh_dp/

$\endgroup$
4
$\begingroup$

Another possibility is Introduction to the Dirichlet Distribution and Related Processes, but I'm afraid I haven't read it yet, however it is down for our next reading group!

$\endgroup$
1
  • $\begingroup$ Looks promising (if long)! $\endgroup$ Commented Nov 29, 2012 at 19:40
1
$\begingroup$

Here's a rather good introductory video lecture by Tom Griffiths, also given at the Machine Learning Summer School.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.