Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: That other flavour...



With these logical semantics, the * and ** interpretations of the Kaucher addition

[-1,1]+[12,8]=[11,9]

are

Q(x,dual[-1,1]), Q(z,[11,9]), Q(y,dual[12,8]), z=x+y

and

Q(y,[12,8]), Q(z,dual[11,9]), Q(x,[-1,1]), z=x+y

isn't it? So the quantifier to be associated with a variable depends on both the * or ** interpretation, and the type of variables. This is where the "logical semantics" is confusing. It is much better to keep only one type of interpretation, e.g. the * one, all ** interpretations being equivalent up to Kaucher dual.

Note that Mr. Hayes' document does not present modal intervals at all (a modal interval is a pair made of a classical interval and a quantifier, this definition being not given in the document).

Alexandre




On Tue, Sep 24, 2013 at 3:24 PM, Michel Hack <mhack@xxxxxxx> wrote:
Alexandre Goldsztejn a écrit:
> >   Then  proper([l,r])  means  Ex(x in Set([l,r]))    (l <= r)
> >         improp([l,r])  means  Ax(x in Set([l,r]))    (l >= r)
>
> This is a common misunderstanding of the modal intervals theory.

Yes, I left out something important:  the quantification applies
to some specific *property* P(x) of elements of the set, not to
the existence itself.  I should have written:

   Prop(X)  is associated with   Ex(x \in Set(X) : P(x))

(That's what happens when quickly scribbled notes are transferred
to a more formal document such as a posting.  Sorry.)

Nate explained it more clearly in his section 1.1 "Logical Semantics".

Michel.
---Sent: 2013-09-24 13:37:00 UTC



--
Dr. Alexandre Goldsztejn

CNRS - Laboratoire d'Informatique de Nantes Atlantique
Office : +33 2 51 12 58 37 Mobile : +33 6 78 04 94 87
Web: www.goldsztejn.com
Email: alexandre.goldsztejn@xxxxxxxxxxxxxx