[ACT-R-users] approximate floating point retrieval
Alexander Petrov
apetrov at uci.edu
Thu Dec 4 13:47:43 EST 2003
Myriam,
These "floating-point values"are not included in standard ACT-R, as far as
I know. I call them "magnitudes" and use them extensively in an ACT-R
model of psychophysical scaling called ANCHOR. Please email me directly
(apetrov at uci.edu) if you are interested in references or preprints. My
code, however, is developed outside the ACT-R implementation and is not
written in Lisp.
Within the standard Lisp implementation, your best bet probably is to
extend (or redefine?) ACT-R's similarity function and turn partial matching
on. I've lost track of the details but I'm sure other people on this list
can advise.
I'm curious what your application is? I mean, in what circumstances do you
think such floating-point values are necessary.
Hope that helps,
-- Alexander Petrov
At 08:15 AM 12/4/2003, contractor wrote:
>How to you retrieve a chunk with an approximate floating point value?
>Thanks.
>
>Here's an example code that gives me an error when not retrieving the
>exact value.
>
>(clear-all)
>
>(chunk-type goal state)
>(chunk-type mynum val)
>
>(add-dm (mygoal isa goal state first)
> (mynum0 isa mynum val 1.48)
> (mynum1 isa mynum val 1.54)
> (mynum2 isa mynum val 2.05)
> (mynum3 isa mynum val 3.5555)
> (mynum4 isa mynum val 3.68))
>
>(p one
> =goal>
> isa goal
> state first
> ==>
> +retrieval>
> isa mynum
> val 3.55
> =goal>
> state second
>
>
>(p two
> =goal>
> isa goal
> state second
> =retrieval>
> isa mynum
> val =val
> ==>
> !output! (=val)
> =goal>
> state stop)
>
>(goal-focus mygoal)
>
>(sgp :er t :v t)
>
>
>
>
>--
>
>Myriam Abramson, Ph.D.
>ITT Industries, Inc.
>202-404-7342
>
>
>_______________________________________________
>ACT-R-users mailing list
>ACT-R-users at act-r.psy.cmu.edu
>http://act-r.psy.cmu.edu/mailman/listinfo/act-r-users
More information about the ACT-R-users
mailing list