PGM Class

[Dave Colwell]

     John and I are going to take the Stanford class on Probabilistic Graphical Models, which begins on September 24.  We've ordered the professor's book by the same title as the class.  Although the book is not required to complete the class it provides more detail and breadth.  This is a graduate level class and is apparently considered challenging even to Stanford grad students.  It's very much in line with the AI and Machine Learning classes we've already taken and should deepen our understanding of modelling and decision making as we'll be expected to code solutions in Octave each week.  Please join us if you can - the more minds the merrier!

Dave

[johnh]

The book just arrived today and it is a BRICK--1200+ pages. Dave and I decided to purchase it so we could get a head start on the lectures. If you have no interest in that, I'd suggest waiting to buy until you can get a look at one of our copies. As Dave says, it's not required for the class. And there's no way all the material will be covered. You should be able to do just fine in the course without the book.

I am in no way bad-mouthing the text. It appears to be very thorough. I haven't sampled the writing, but if Professor Koller can write as well as she lectures, it should be very clear. I only suggest that it is probably more than you need to get through the course.