Humanist Discussion Group, Vol. 38, No. 224.
        Department of Digital Humanities, University of Cologne
                      Hosted by DH-Cologne
                       www.dhhumanist.org
                Submit to: humanist@dhhumanist.org




        Date: 2024-11-04 13:28:38+00:00
        From: Manfred Thaller <manfred.thaller@uni-koeln.de>
        Subject: AI and ML

Dear Humanists,

as quite a few of the readers of this list seem to have an interest in
AI - and therefore in the recent claims of the relevance of LMMs and ML
in general for it - I would like to point to a recent publication. My
thanks to Michael Piotrowski for mentioning it to me. (If it has already
been discussed in Humanist: my apologies to everyone.)

At https://link.springer.com/article/10.1007/s42113-024-00217-5 you find
in the context of cognitive science a claim that the proof submitted
there shows, that it is inherently impossible to base an Artificial
General Intelligence upon ML. Or not impossible, but rather "formally
NP-hard; i.e. possible in principle but provably infeasible;"

While complexity theory is not a part of computer science I feel
particularly comfortable with, I am unable to find a fault in that proof.

That may be of interest to parts of the community trying to evaluate the
importance and potential of ML and LMMs.

Best regards,
Manfred


_______________________________________________
Unsubscribe at: http://dhhumanist.org/Restricted
List posts to: humanist@dhhumanist.org
List info and archives at at: http://dhhumanist.org
Listmember interface at: http://dhhumanist.org/Restricted/
Subscribe at: http://dhhumanist.org/membership_form.php