I don't recall seeing that implemented in Racket/Scheme, but, in class, years ago, Leslie Kaelbling mentioned using Scheme captured continuations for AI search backtracking, as I mentioned (and Matthias has good comments in that thread): https://groups.google.com/d/msg/racket-users/jHPtw3Gzqk4/AAqsc-x-AgAJ

That might've been in the class for which we were using a draft of the Russell&Norvig text; I don't know whether it was mentioned in there.  IIRC, Leslie at the time was coming from Stanford AI Lab tradition (West Coast tradition OG, to MIT's East Coast).

For most purposes, perhaps one would probably want to write a search two different ways: one that takes advantage of Scheme's first-class continuations, and one that doesn't; and compare them (both performance and ease of implementation).

I don't know how relevant to performance it would be that we almost never see first-class continuations being leveraged directly in users' code (outside of the implementation of a Scheme itself), and how that might have affected priorities in the Scheme implementation.  The implementor of a particular Scheme could say.

--
You received this message because you are subscribed to the Google Groups "Racket 
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/racket-users/4ed3134c-54fd-5747-486d-c45ce93a99bb%40neilvandyke.org.

Reply via email to