Why Probabilistic Programming Matters

Re:Why Probabilistic Programming Matters by JadeNB on Tuesday July 15, 2014 @06:03PM (#8039355)

Every time that I see a discussion of probabilistic programming, I'm reminded that I want to try to find an excuse to use IBAL (http://www.gelberpfeffer.net/avi.htm). I used to have a binary for it lying around, but can't find it any more; does anyone know where to get it?

[ Parent | on HN ]

Re:Why Probabilistic Programming Matters by davmre on Tuesday July 15, 2014 @08:16PM (#8039923)

IBAL is quite old; you're probably better off learning something that's still being actively maintained. The natural choice would be Avi Pfeffer's new language, Figaro (https://www.cra.com/commercial-solutions/probabilistic-model...).

[ Parent | on HN ]

Re:Why Probabilistic Programming Matters by JadeNB on Tuesday July 15, 2014 @08:52PM (#8040022)

Thank you! I will have a look at it. Do you know if there's any analogue of the 99 Haskell / Lisp / Prolog problems list geared toward probabilistic programming?

[ Parent | on HN ]

Re:Why Probabilistic Programming Matters by davmre on Wednesday July 16, 2014 @05:33AM (#8041264)

Not as far as I know. The closest I can think of is Josh Tenenbaum and Noah Goodman's online book "Probabilistic Models of Cognition", https://probmods.org, which has lots of examples of simple models, written in Church; these could be a good set of exercises to try translating into Figaro. Though these systems are immature enough that there's no guarantee Figaro will actually work on the translated models (or that Church will even work on the originals :-).

[ Parent | on HN ]