On Thu, Dec 2, 2021 at 7:24 AM 'Nil Geisweiller' via opencog <
[email protected]> wrote:

> Hi Patrick!
>
> I think OpenCog and NARS differ a bit here. [...] I'm pretty much blank
> slate on the subject
>
>
NARS in it's earliest conception was a paper by Pei Wang that described
formulas for reasoning, not unlike the PLN formulas, but differing in
details.  Part of the vision for URE was to allow NARS (or other kinds of
deductive/inductive systems) to be implemented, as "just another set of
rules".

I think this is still the case today. You could "easily" implement NARS
today, by cloning the PLN repo, gutting 95% of the rules in it,  keeping
around a couple as examples, and then start adding the NARS rules. It
should be "easy" with an appropriate definition of "easy".

Now Ben didn't like NARS, back in the day, because some of the rules
appeared to violate conventional Bayesian probability, and he figured his
bread was buttered on the side closer to conventional probability theory.
Thus PLN. But anyway, with the URE, you can experiment with any kind of
deductive system you want.

These days, I don't like NARS or PLN, because I think I know how to learn
common-sense reasoning "from scratch". That includes automatically learning
PLN-like or NARS-like rules, whatever they may be, as appropriate to
whatever setting they've been learned for.  Doing this is on my TODO list,
but  at the current rate of progress, this is at least a decade away.

--linas

-- 
Patrick: Are they laughing at us?
Sponge Bob: No, Patrick, they are laughing next to us.

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/opencog/CAHrUA356X0upDHw%3DdfXsA%3DYLc_FeTwTHRUrqF6QFrE3P-kV8iQ%40mail.gmail.com.

Reply via email to