On 10/20/2016 04:53 PM, Nil Geisweiller wrote:
then we can normalize the positive-evidence in the obvious way (i.e.
the total_evidence we want for the InheritanceLink is the total number
of entities satisfying the SatisfyingSet, within the assumed overall
universe of discourse)...
No, it is the count of (SetLink (Concept "Ben")), which is 1. Or if that
weird M2I as currently implemented rule is correct, it is the count of
(Concept "Ben"). Do we agree on that?
Actually it would be at most 1 (or the size of (Concept "Ben") assuming
the M2I rule I don't understand), since we need to account for the
uncertainty on the conditional probability itself.
In the PLN book it is confidence over the MemberLink times some constant
which depends of the cohesiveness of the set (here the satisfying set of
eat(X, cat)). I do see the stuff, but it's not just about cohesiveness,
there is another implicit assumption, which is that
if X and Y are in S, then Set X Y are in S, then the whole stuff starts
making sense, but cohesiveness alone is not enough to capture that, I
think. We already had an email exchange about this I'm sure, but I
forgot the conclusion (if there were any).
Nil
If this approach is right, then the PLN formulas like Eval2Member and
M2I would propagate positive_evidence values rather than strengths...
Since I can't make sense of these rules and I'm not sure what they are
supposed to be, I can't comment.
No changes for the PLN rules acting on straightforwardly probabilistic
links like InheritanceLinks would be needed, because these rules use
strength which can be derived from positive_evidence and
total_evidence...
Agreed.
Adding new TV types like this is annoying, but OTOH abusing the
"strength" or "mean" field in an existing TV type by making it mean
positive_evidence for certain link types, seems even more annoying...
Agreed. Ultimately I think we could offer a positive_evidence setter.
But I still don't understand the need for propagating positive_evidence
without knowing the total count.
Nil
Hopefully someone will bite the bullet and implement ProtoAtom soon ;p ;)
thoughts?
ben
--
You received this message because you are subscribed to the Google Groups
"opencog" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/opencog.
To view this discussion on the web visit
https://groups.google.com/d/msgid/opencog/5808D2BB.1020204%40gmail.com.
For more options, visit https://groups.google.com/d/optout.
