So I found myself surfing the tubes, looking for code examples of Bayes networks written in Lisp to compliment my recent studies for Stanford’s online ‘Introduction to Artificial Intelligence’ course… because that’s what I do for fun.
After a fair amount of digging I’ve come across some Lisp code from one of Stuart Russell’s A.I. classes at Berkeley that I’d like to share with you, via the way-back machine.
Interestingly I found more than a few files in the catacombs that aren’t in the main repository or possibly very different. Some nuggests include bayes-nets.lisp, elimination.lisp, probability.lisp, sampling.lisp, and statistics.lisp.
Eureka! I’ve found the ancient artifacts and my journey is near completion. However I was haunted by my own conspiracies of why they were lost to the internet archive to begin with.
So, recalling Occam’s Razor, I decided to mail Peter Norvig and see what he had to say about it. This was his reply. ‘Yes, Stuart did some code for various versions of the course, and he meant to integrate them bacj into the main branch but never did…’