anybody who knows how to do variable elimination?

Question

anonymous · Answer

What kind of variable elimination? Surely, a masters student in CS knows how to solve systems of equations? :D

anonymous · Answer

Hehe, well it depends on what kinds of equations :) I'm not strong with probability and bayesian networks at all. I'm having difficulties understanding the variable elimination procedure for inference in bayesian networks. Especially the part on how factors are created, and manipulated. So any help on this would be greatly appreciated :)

anonymous · Answer

Well, the way I do it is by using a table similar to Karnaugh maps to minimize the logic. I'm not an expert on Bayesian networks though I think this paper will find you well if you haven't seen it already: http://www.cs.uiuc.edu/class/sp08/cs440/notes/varElimLec.pdf

anonymous · Answer

Thanks, hadn't thought about Karnaugh maps ... that was a good idea :) I've read it, but I'll try again. Thanks for the feedback :)

anonymous · Answer

Yeah, unfortunately, I'm an undergrad in EE/Applied Math, so some of my insight for CS comes from EE. That said, I haven't strictly learned about Bayesian networks, so I probably won't be much help to you. Hopefully someone else sees this and answers it. I'd recommend you ask the question as well on the main openstudy site.

anonymous · Answer

Oh, I just joined the site. Didn't realize there were these sections, I just posted, but I think use of Karnaugh maps could really help me out here, so I'll look into that first before asking more questions ;)