XPost: sci.physics
From: janburse@fastmail.fm
Hi,
Maybe a probem with the Logic Programming community
is the fact, that they are totally clueless about
linear algebra. Something tells me, we I could
compute the transitive closure via matrix inversion:
(I-M)^(-1) = I + M + M^2 + M^3 + ...
I am pretty sure there are a few Lean or even Rocq
theorems already formalized, showing some of these
identities and their numerical perks.
Still I find here long discussions (**), mareveling
how to compute the transitive closure, in a
sequential setting. But already when M is sparse
things are pretty speedy with linear algebra. Matrices
are perfect for AI Accelerators that are specialized
for linear algebra, since it is one of the most
parallelizable computations in numerical linear algebra.
GPU and distributed HPC performance is excellent,
but one has possibly to consider AI Accelerators nowadays
on laptop and micro servers, for so called Local AI.
Bye
P.S.: (**)
Wanna see a bunch of clueless people associated to Stony Brook or not:
https://discourse.prolog-lang.org/t/subgraph-isomorphism-prolog-
s-conventional/173/7
P.P.S.: I am currently looking into combinatorial search,
but transitive closure is of course also a candidate to
delegate to an AI accelerator when doing Local AI.
Mild Shock schrieb:
> Hi,
>
> Woa! I could have told you so. Although hired
> by EricGT to provoke me on SWI-Prolog discourse,
> and then EricGT kick me out of SWI-Prolog discourse,
>
> on grounds "not listening to staff". The same
> Julio Di Egidio who invented nonsense stuff on
> SWI-Prolog discouse, to annoy me.
>
> Had his lesson teached about the usual code of
> conduct in politically correct forums, where one
> has to adopt a "performative" way of telling
>
> technical thruths, either by not telling them,
> because the community has already lost their
> moral compas, and is only producing hot air,
>
> or just quitting the community. The lesson was
> already posted in 2023:
>
> Théo Zimmermann
> For anyone who might have followed this discussion or any of the recent
> discussions in which Julio Di Egidio was a participant and which turned
> sour, we (administrators of the Coq Zulip and members of the Code of
> Conduct enforcement team) want to let everyone know that we do not
> consider this behavior acceptable (regardless of its technical content).
> A prior warning had been addressed to Julio in August 2023. Following
> this warning and recent new reports, we have decided to send a new
> warning accompanied by a temporary ban, that could be lifted with a
> statement that the offender would follow our rules in the future.
>
> Bye
>
> Julio Di Egidio schrieb:
> > On 10/07/2025 14:31, Julio Di Egidio wrote:
> >> On 10/07/2025 14:01, Julio Di Egidio wrote:
> >>> The end of the world is nigh:
> >>>
> >>>
>
>
> >>>
> >>> From interactive proof assistant to completely upside-down and
> >>> completely broken, and not just on that at that point of course.
> >>>
> >>> And the fucking shamelessness...
> >>
> >> But we must thank MS for the nail in that coffin, too: they can't
> >> be satisfied with just a Lean broken by design, they must own the
> >> whole compartment: only poisoned meatballs for the public...
> >
> > Follow up on Rocq and the appointed pieces of ungodly shit:
> >
> > Now if I try to login:
> >
> > << Account suspended: No constructive purpose to their actions other
> than creating dissent within the community. >>
> >
> > You pieces of FRAUDULENT UNGODLY SHIT, the defamation added to the
> > abuse, abuse of power as well as personal.
> >
> > And the last blow, just to mention the icing on the cake, if the
> > great people they have been onboarding:
> >
> > << FYI: I have no animosity towards you. >>
> > (Dear Thomas I-am-a-retarded-Nazi Lamiaux, FYI I don't give a shit
> > whether you smile or not while shitting all over the place, you
> > fucking nazi-retarded pieces of shit, you and the whole gang...)
> >
> > YOU UNGODLY FRAUDULENT AND RETARDED PIECES OF SHIT, now it's all
> > Microsoft and Lean and Lean 2 and sub-Lean and proto-Lean... with
> > Microsoft's world-famous best practices, top guidance and impeccable
> > products tailored to their customers' real needs...
> >
> > Indeed you managers and maintainers and marketers and the fucking
> > corporate ladder and the whole fucking pyramid indeed: on behalf
> > of real humanity, I wish you all a rapid ass cancer!!
> >
> > Bottom line:
> >
> > Fuck you. And fuck you especially for 20 years of Coq that you have
> > managed to destroy in less than 5 years, you pieces of pyramidic shit.
> >
> > *Plonk*
> >
> > Julio
>
>
> Mild Shock schrieb:
>> Mild Shock schrieb:
>>> Concerning this boring nonsense:
>>>
>>> https://book.simply-logical.space/src/text/2_part_ii/5.3.html#
>>>
>>> Funny idea that anybody would be interested just now in
>>> the year 2025 in things like teaching breadth first
>>> search versus depth first search, or even be “mystified”
>>> by such stuff. Its extremly trivial stuff:
>>>
>>> Insert your favorite tree traversal pictures here.
>>>
>>> Its even not artificial intelligence neither has anything
>>> to do with mathematical logic, rather belongs to computer
>>> science and discrete mathematics which you have in
>>> 1st year university
>>>
>>> courses, making it moot to call it “simply logical”. It
>>> reminds me of the idea of teaching how wax candles work
>>> to dumb down students, when just light bulbs have been
>>> invented. If this is the outcome
>>>
>>> of the Prolog Education Group 2.0, then good night.
>>
>
--- SoupGate-Win32 v1.05
* Origin: you cannot sedate... all the things you hate (1:229/2)
|
