Re: Heuristic backtrack in Prolog?

On Jun 11, 3:42 pm, "Neng-Fa Zhou" <n...@xxxxxxx> wrote:
Please take a look at PRISM (http://sato-www.cs.titech.ac.jp/prism/), which
is an extension of Prolog with support of probabilistic reasoning and
learning. You can not only do probabilistic parsing using given probability
distributions, but also learning to estimate probability distributions from
given data. I am sure your life will be much easier if you use PRISM instead
of Scheme.

Cheers,
Neng-Fa

Neng-Fa,

Could you please give a brief example of how to do this in PRISM? I
had a look at it, and tried out the 'viterbig' predicate to unify with
the most likely goal. However, this does not back track onto the
second most likely, third most... I think Panu was after a greedy
search that follows the most likely first rather than a probabilistic
search that throws a dice on each search node based on the
probabilities of each branch.

Panu,

There is an example here:

http://www.csupomona.edu/~jrfisher/www/prolog_tutorial/5_1.html

That shows how to implement an A* heuristic search in prolog. Prolog
only provides depth first search, so you have to explicitly adapt that
mechanism with a program to explore different orderings, roughly
speaking by storing the alternatives in a list ordered by the
heuristics. So its not too hard to write a greedy search that follows
the highest probability first, using a similar mechansim to this
example. You could calculate the probabilities yourself in standard
prolog, or use PRISM to do the probability calculations for you.

Rupert

.

Relevant Pages

  • Heuristic backtrack in Prolog?
    ... I'm trying to learn Prolog because it's a very interesting language and seems to be on par with LISP with respect to freedom of expression:) My own experience with Prolog is limited to writing some parsers and recursive predicates such as permutations and so on. ... Another execution environment I've needed is a heuristic parser where each alternative parse has a probability. ... Ordinary Prolog backtracking is a special case of such an environment where fail:- probability. ...
    (comp.lang.prolog)
  • Re: Heuristic backtrack in Prolog?
    ... which is an extension of Prolog with support of probabilistic reasoning and learning. ... You can not only do probabilistic parsing using given probability distributions, but also learning to estimate probability distributions from given data. ... because I like the elegance of Prolog a lot. ...
    (comp.lang.prolog)
  • Re: Heuristic backtrack in Prolog?
    ... Another execution environment I've needed is a heuristic parser where each ... alternative parse has a probability. ... do I write the probability/1 predicate in Prolog, ...
    (comp.lang.prolog)
  • Re: So called "stimulus/response" models
    ... "1) Most learning, as far as I know, maybe even all learning, can be reduced ... to "unsupervised learning" of probability distributions (to use the machine ... but when cognitions are in conflict ... Perception and Neural Function". ...
    (comp.ai.philosophy)
  • Re: Probability of picking a positive rational number at random
    ... Classical probability is more limited than measure-theoretic ... there are infinitely many distributions satisfying that relation. ... distributions over all well-orderings of distributions over natural ... natural numbers, in a bijection to the unit interval of reals, by ...
    (sci.math)